Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

🚀 Introducing GenLit – Reformulating Single-Image Relighting as Video Generation! We leverage video diffusion models to perform realistic near-field relighting from just a single image—No explicit 3D reconstruction or ray tracing required! No intermediate graphics buffers, directly in the pixel space! 📄 Dive into the paper: 🎥 Project page... show more

Haven Feng

2,199 subscribers

22,442 views • 1 year ago •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Selected as a best paper finalist at #CVPR2026: PixelDiT from NVIDIA Research In most image generation models, a pretrained autoencoder compresses the image before any diffusion happens, causing quality loss that accumulates across the entire pipeline. PixelDiT, or Pixel Diffusion Transformers, removes this step entirely. It's a single-stage model that learns the diffusion process directly in pixel space, end-to-end.

Selected as a best paper finalist at #CVPR2026: PixelDiT from NVIDIA Research In most image generation models, a pretrained autoencoder compresses the image before any diffusion happens, causing quality loss that accumulates across the entire pipeline. PixelDiT, or Pixel Diffusion Transformers, removes this step entirely. It's a single-stage model that learns the diffusion process directly in pixel space, end-to-end.

NVIDIA AI

28,129 views • 1 month ago

Wonderland: Navigating 3D Scenes from a Single Image Contributions: • First, we introduce a representation for controllable 3D generation by leveraging the generative priors from camera-guided video diffusion models. Unlike image models, video diffusion models are trained on extensive video datasets. This enables them to capture comprehensive spatial relationships within scenes across multiple views and embed a form of "3D awareness" in their latent space, which allows us to maintain 3D consistency in novel view synthesis. • Second, to achieve controllable novel view generation, we empower video models with precise control over specified camera motions. We introduce a novel dual-branch conditioning mechanism that effectively incorporates desired diverse camera trajectories into the video diffusion model. This enables expansion of a single image into a multi-view consistent capture of a 3D scene with precise pose control. • Third, to achieve efficient 3D reconstruction, we directly transform video latents into 3DGS. We propose a novel latent-based large reconstruction model (LaLRM) that lifts video latents to 3D in a feed-forward manner. With this design, during inference, our model directly predicts 3DGS from a single input image, effectively aligning the generation and reconstruction tasks—and bridging image space and 3D space—through the video latent space. Compared with reconstructing scenes from images, the video latent space offers a 256× spatial-temporal reduction while retaining essential and consistent 3D structural details. Such a high degree of compression is crucial, as it allows the LaLRM to handle a wider range of 3D scenes within the reconstruction framework, with the same memory constraints.

Wonderland: Navigating 3D Scenes from a Single Image Contributions: • First, we introduce a representation for controllable 3D generation by leveraging the generative priors from camera-guided video diffusion models. Unlike image models, video diffusion models are trained on extensive video datasets. This enables them to capture comprehensive spatial relationships within scenes across multiple views and embed a form of "3D awareness" in their latent space, which allows us to maintain 3D consistency in novel view synthesis. • Second, to achieve controllable novel view generation, we empower video models with precise control over specified camera motions. We introduce a novel dual-branch conditioning mechanism that effectively incorporates desired diverse camera trajectories into the video diffusion model. This enables expansion of a single image into a multi-view consistent capture of a 3D scene with precise pose control. • Third, to achieve efficient 3D reconstruction, we directly transform video latents into 3DGS. We propose a novel latent-based large reconstruction model (LaLRM) that lifts video latents to 3D in a feed-forward manner. With this design, during inference, our model directly predicts 3DGS from a single input image, effectively aligning the generation and reconstruction tasks—and bridging image space and 3D space—through the video latent space. Compared with reconstructing scenes from images, the video latent space offers a 256× spatial-temporal reduction while retaining essential and consistent 3D structural details. Such a high degree of compression is crucial, as it allows the LaLRM to handle a wider range of 3D scenes within the reconstruction framework, with the same memory constraints.

MrNeRF

52,849 views • 1 year ago

Introducing 📦𝗔𝗿𝘁𝗶𝗟𝗮𝘁𝗲𝗻𝘁🔧 (SIGGRAPH Asia 2025) — a high-quality 3D diffusion model that explicitly models object articulation, paving the way for richer, more realistic assets in embodied AI and simulation: – Generates fully articulated 3D objects – Physically plausible joints & motion – High-fidelity 3D Gaussian appearance – Supports generation from a single real image arXiv: Project: Code (coming soon):

Introducing 📦𝗔𝗿𝘁𝗶𝗟𝗮𝘁𝗲𝗻𝘁🔧 (SIGGRAPH Asia 2025) — a high-quality 3D diffusion model that explicitly models object articulation, paving the way for richer, more realistic assets in embodied AI and simulation: – Generates fully articulated 3D objects – Physically plausible joints & motion – High-fidelity 3D Gaussian appearance – Supports generation from a single real image arXiv: Project: Code (coming soon):

Xingang Pan

11,517 views • 8 months ago

Create a 3D model from a single image, set of images or a text prompt in < 1 minute 😮‍💨 This new AI paper called CAT3D shows us that it’ll keep getting easier to produce 3D models from 2D images — whether it’s a sparser real world 3D scan (a few photos instead of hundreds) or your favorite 2D image generator like Midjourney (just an image). How does this magic work? “This architecture is similar to video diffusion models, but with camera pose embeddings for each image instead of time embeddings. The generated views are passed into a robust 3D reconstruction pipeline to create the 3D representation (Zip-NeRF or 3DGS)”

Create a 3D model from a single image, set of images or a text prompt in < 1 minute 😮‍💨 This new AI paper called CAT3D shows us that it’ll keep getting easier to produce 3D models from 2D images — whether it’s a sparser real world 3D scan (a few photos instead of hundreds) or your favorite 2D image generator like Midjourney (just an image). How does this magic work? “This architecture is similar to video diffusion models, but with camera pose embeddings for each image instead of time embeddings. The generated views are passed into a robust 3D reconstruction pipeline to create the 3D representation (Zip-NeRF or 3DGS)”

Bilawal Sidhu

92,792 views • 2 years ago

Video diffusion models have strong implicit representations of 3D shape, material, and lighting, but controlling them with language is cumbersome, and control is critical for artists and animators. GenLit connects these implicit representations with a continuous 5D control signal describing the direction and intensity of a point light source. This enables single-image near-field relighting of an image using a video diffusion model. We use a ControlNet-like approach and show that, with a small amount of synthetic data, GenLit generalizes to complex real-world images. Given a single image and the 5D lighting signal, GenLit creates a video of a moving light source that is inside the scene. It moves around and behind scene objects, producing effects such as shading, cast shadows, secularities, and interreflections with a realism that is hard to obtain with traditional inverse rendering methods. GenLit shows that it is possible to get continuous control over implicit physical processes within a video model. I think this is just the beginning and promises to make such models much more practical for creators. Shrisha Bharadwaj will present today at SIGGRAPH Asia Room: S423/S424, Level 4 @ 13:50 on 15 of Dec.

Video diffusion models have strong implicit representations of 3D shape, material, and lighting, but controlling them with language is cumbersome, and control is critical for artists and animators. GenLit connects these implicit representations with a continuous 5D control signal describing the direction and intensity of a point light source. This enables single-image near-field relighting of an image using a video diffusion model. We use a ControlNet-like approach and show that, with a small amount of synthetic data, GenLit generalizes to complex real-world images. Given a single image and the 5D lighting signal, GenLit creates a video of a moving light source that is inside the scene. It moves around and behind scene objects, producing effects such as shading, cast shadows, secularities, and interreflections with a realism that is hard to obtain with traditional inverse rendering methods. GenLit shows that it is possible to get continuous control over implicit physical processes within a video model. I think this is just the beginning and promises to make such models much more practical for creators. Shrisha Bharadwaj will present today at SIGGRAPH Asia Room: S423/S424, Level 4 @ 13:50 on 15 of Dec.

Michael Black

22,182 views • 7 months ago

DimensionX: Create Any 3D and 4D Scenes from a Single Image with Controllable Video Diffusion TL;DR: Create 3/4DGS from Video Diffusion Note: Some first inference code released (not all yet). Contributions (cited): • We present DimensionX, a novel framework for generating photorealistic 3D and 4D scenes from only a single image using controllable video diffusion. • We propose ST-Director, which decouples the spatial and temporal priors in video diffusion models by learning (spatial and temporal) dimension-aware modules with our curated datasets. We further enhance the hybriddimension control with a training-free composition approach according to the essence of video diffusion denoising process. • To bridge the gap between video diffusion and real-world scenes, we design a trajectory-aware mechanism for 3D generation and an identity-preserving denoising approach for 4D generation, enabling more realistic and controllable scene synthesis. • Extensive experiments manifest that our DimensionX delivers superior performance in video, 3D, and 4D generation compared with baseline methods.

DimensionX: Create Any 3D and 4D Scenes from a Single Image with Controllable Video Diffusion TL;DR: Create 3/4DGS from Video Diffusion Note: Some first inference code released (not all yet). Contributions (cited): • We present DimensionX, a novel framework for generating photorealistic 3D and 4D scenes from only a single image using controllable video diffusion. • We propose ST-Director, which decouples the spatial and temporal priors in video diffusion models by learning (spatial and temporal) dimension-aware modules with our curated datasets. We further enhance the hybriddimension control with a training-free composition approach according to the essence of video diffusion denoising process. • To bridge the gap between video diffusion and real-world scenes, we design a trajectory-aware mechanism for 3D generation and an identity-preserving denoising approach for 4D generation, enabling more realistic and controllable scene synthesis. • Extensive experiments manifest that our DimensionX delivers superior performance in video, 3D, and 4D generation compared with baseline methods.

MrNeRF

17,052 views • 1 year ago

📢Pix2NPHM: Learning to Regress NPHM Reconstructions From a Single Image📢 We directly regress neural parametric head models (NPHMs) from a single image — fast, stable, and significantly more expressive than classical 3DMMs such as FLAME. Face tracking & 3D reconstruction are often limited by the representational capacity of PCA-based face models. By lifting NPHMs to a first-class reconstruction primitive, we enable more accurate geometry, richer expressions, and finer animation control. Pix2NPHM obtains fast and reliable NPHM reconstructions on real-world data. Inference-time optimization against surface normals and canonical point maps can further increase fidelity. Key to successful and generalized training of our ViT-based network are: (1) large-scale registration of existing 3D head datasets, and (2) self-supervised training on vast in-the-wild 2D video datasets using pseudo ground-truth surface normals. Finally, we show that geometry-aware pretraining on pixel-aligned reconstruction tasks significantly outperforms generic visual pretraining (e.g., DINO-style features) in terms of generalization. 🌍 🎥 Great work by Simon Giebenhain, Tobias Kirschstein, Liam Schoneveld, Davide Davoli, Zhe Chen

📢Pix2NPHM: Learning to Regress NPHM Reconstructions From a Single Image📢 We directly regress neural parametric head models (NPHMs) from a single image — fast, stable, and significantly more expressive than classical 3DMMs such as FLAME. Face tracking & 3D reconstruction are often limited by the representational capacity of PCA-based face models. By lifting NPHMs to a first-class reconstruction primitive, we enable more accurate geometry, richer expressions, and finer animation control. Pix2NPHM obtains fast and reliable NPHM reconstructions on real-world data. Inference-time optimization against surface normals and canonical point maps can further increase fidelity. Key to successful and generalized training of our ViT-based network are: (1) large-scale registration of existing 3D head datasets, and (2) self-supervised training on vast in-the-wild 2D video datasets using pseudo ground-truth surface normals. Finally, we show that geometry-aware pretraining on pixel-aligned reconstruction tasks significantly outperforms generic visual pretraining (e.g., DINO-style features) in terms of generalization. 🌍 🎥 Great work by Simon Giebenhain, Tobias Kirschstein, Liam Schoneveld, Davide Davoli, Zhe Chen

Matthias Niessner

37,850 views • 7 months ago

Google dropped a new AI paper called LUMIERE. It's remarkably flexible, supporting video inpainting, image-to-video, AND stylized video generation tasks. Say hello to “space-time diffusion” for video generation! Now what the heck does that mean exactly?! 🌐⏳ → TL;DR it utilizes a “Space-Time UNet” architecture that generates the full duration of the video in one pass, rather than generating distant keyframes and interpolating between them like prior works. Because the computation is done in this “compressed space-time representation” to generate the full clip at once, it's far more temporally consistent. → Another benefit of generating the full video at once is that you can “direct” the video generation, making it easier to hand off to other models/tasks without having to stitch together partial solutions. You can condition generations on additional inputs, meaning you get the full stack of AI video capabilities – from video inpainting to image-to-video and beyond. → New SOTA for AI video generation? User study results in the paper suggest human evaluators preferred Lumiere over Runway Gen-2, Pika Labs, and Stable Video Diffusion in terms of quality, text alignment AND motion. But as always, we need to get hands-on with this tech when Google *actually* decides to ship it. → Could this end up inside YouTube? Y’all know i’m obsessed with blending reality and imagination – so it’s the video inpainting tech I'm most excited about. I really hope this model finds its way into YouTube's Generative AI efforts, and based on their prior announcements and the list of acknowledgments in the paper I think it might! 🤞🏽 Links: 🔗Paper: 🔗Project:

Google dropped a new AI paper called LUMIERE. It's remarkably flexible, supporting video inpainting, image-to-video, AND stylized video generation tasks. Say hello to “space-time diffusion” for video generation! Now what the heck does that mean exactly?! 🌐⏳ → TL;DR it utilizes a “Space-Time UNet” architecture that generates the full duration of the video in one pass, rather than generating distant keyframes and interpolating between them like prior works. Because the computation is done in this “compressed space-time representation” to generate the full clip at once, it's far more temporally consistent. → Another benefit of generating the full video at once is that you can “direct” the video generation, making it easier to hand off to other models/tasks without having to stitch together partial solutions. You can condition generations on additional inputs, meaning you get the full stack of AI video capabilities – from video inpainting to image-to-video and beyond. → New SOTA for AI video generation? User study results in the paper suggest human evaluators preferred Lumiere over Runway Gen-2, Pika Labs, and Stable Video Diffusion in terms of quality, text alignment AND motion. But as always, we need to get hands-on with this tech when Google actually decides to ship it. → Could this end up inside YouTube? Y’all know i’m obsessed with blending reality and imagination – so it’s the video inpainting tech I'm most excited about. I really hope this model finds its way into YouTube's Generative AI efforts, and based on their prior announcements and the list of acknowledgments in the paper I think it might! 🤞🏽 Links: 🔗Paper: 🔗Project:

Bilawal Sidhu

44,822 views • 2 years ago

You can't 3D reconstruct glass from images... ...WRONG! Thanks for video diffusion, now just about anything is possible! Introducing...Diffusion Knows Transparency (DKT) Transparent and reflective objects usually break robot vision and photogrammetry pipelines because they don't follow the "solid object" rules standard cameras expect. DKT is a new AI model that repurposes the "internal physics engine" found in video generation models to solve this problem. Researchers took a massive video diffusion model (WAN) and fine-tuned it using a custom-built synthetic dataset to turn it into a high-precision depth sensor. To train the AI, they built the first massive synthetic video library of transparent objects, 1.32 million frames of perfectly labeled glass and metal objects in motion. Without ever seeing a "real" labeled video of glass during training, the model (DKT) outperformed all previous specialized systems on real-world benchmarks (ClearPose, DREDS). They created a "lightweight" 1.3B parameter version that runs fast enough (0.17s per frame) to be used on actual robot hardware. Two reasons I find this project important: 1. It further proves that synthetic data will be essential for training the next generation vision models. 2. In real-world robotic tests, using DKT's depth maps nearly doubled the success rate of robot arms trying to pick up objects on tricky reflective or translucent surfaces. At home robots will need to interact with these types of objects on a daily basis. Check out the project page here: Code is LIVE! #Computervision #Robotics #AI

You can't 3D reconstruct glass from images... ...WRONG! Thanks for video diffusion, now just about anything is possible! Introducing...Diffusion Knows Transparency (DKT) Transparent and reflective objects usually break robot vision and photogrammetry pipelines because they don't follow the "solid object" rules standard cameras expect. DKT is a new AI model that repurposes the "internal physics engine" found in video generation models to solve this problem. Researchers took a massive video diffusion model (WAN) and fine-tuned it using a custom-built synthetic dataset to turn it into a high-precision depth sensor. To train the AI, they built the first massive synthetic video library of transparent objects, 1.32 million frames of perfectly labeled glass and metal objects in motion. Without ever seeing a "real" labeled video of glass during training, the model (DKT) outperformed all previous specialized systems on real-world benchmarks (ClearPose, DREDS). They created a "lightweight" 1.3B parameter version that runs fast enough (0.17s per frame) to be used on actual robot hardware. Two reasons I find this project important: 1. It further proves that synthetic data will be essential for training the next generation vision models. 2. In real-world robotic tests, using DKT's depth maps nearly doubled the success rate of robot arms trying to pick up objects on tricky reflective or translucent surfaces. At home robots will need to interact with these types of objects on a daily basis. Check out the project page here: Code is LIVE! #Computervision #Robotics #AI

Jonathan Stephens

17,712 views • 6 months ago

Everyone's sleeping on image-to-3D AI models. They can make your app look incredibly unique, with just a little effort. Here's how. This is my calorie tracker, built in a week with nothing but prompting. Just Claude Code + a couple APIs. The visuals are all AI-generated. I'll be sharing the full workflow + all the crazy technical stuff Claude and I did to make this work, so nobody has to struggle through it like me. Deep dive coming soon! Till then, this is the high-level idea: 1. Get a clean image of the food (or whatever your asset is) - In my app, the user describes foods via text, or attaches images (or both) - If text, an LLM extracts the food description and formats it into a specific prompt I tuned for this design, and we generate an image using Z-Image Turbo through fal - If image, we do the same thing but with FLUX.2 [dev] to edit the user image into our reference design - Originally, both used Google Nano Banana, but switching to open models cut costs and latency a ton 2. Gaussian splatting (2D image → 3D model) - I tried various 2D-to-3D options on fal and ended up with TripoSplat as my preferred balance of speed, cost, latency; this turns an image into a 3D model that looks super high quality (link below) - The app displays the 2D image while our backend generates the 3D splat - We "groom" the splat to reduce size and load time by culling low-opacity/scale points 3. Render efficiently on device Originally, it looked great but ran at 10 FPS. Getting to 120 FPS was a crazy journey. TL;DR: - SwiftUI had to go; it forced us to render each asset in independent MTKViews, which wasn't workable - Instead, we composite every dish into one full-bleed CAMetalLayer using MetalSplatter (link below) - We had to make some optimizations within MetalSplatter's code too, to reduce the overhead of sorting points per render Then I added some finishing touches like the subtle rotation and parallax as they move around. I think it turned out pretty cool :) Overall, this took some effort, but we still got it done in less than a day. Hopefully your agent can follow in the footsteps of mine and do it much faster. Keep an eye out for the bigger writeup, which'll give your agent everything it needs. If you have any questions, drop em below!

Everyone's sleeping on image-to-3D AI models. They can make your app look incredibly unique, with just a little effort. Here's how. This is my calorie tracker, built in a week with nothing but prompting. Just Claude Code + a couple APIs. The visuals are all AI-generated. I'll be sharing the full workflow + all the crazy technical stuff Claude and I did to make this work, so nobody has to struggle through it like me. Deep dive coming soon! Till then, this is the high-level idea: 1. Get a clean image of the food (or whatever your asset is) - In my app, the user describes foods via text, or attaches images (or both) - If text, an LLM extracts the food description and formats it into a specific prompt I tuned for this design, and we generate an image using Z-Image Turbo through fal - If image, we do the same thing but with FLUX.2 [dev] to edit the user image into our reference design - Originally, both used Google Nano Banana, but switching to open models cut costs and latency a ton 2. Gaussian splatting (2D image → 3D model) - I tried various 2D-to-3D options on fal and ended up with TripoSplat as my preferred balance of speed, cost, latency; this turns an image into a 3D model that looks super high quality (link below) - The app displays the 2D image while our backend generates the 3D splat - We "groom" the splat to reduce size and load time by culling low-opacity/scale points 3. Render efficiently on device Originally, it looked great but ran at 10 FPS. Getting to 120 FPS was a crazy journey. TL;DR: - SwiftUI had to go; it forced us to render each asset in independent MTKViews, which wasn't workable - Instead, we composite every dish into one full-bleed CAMetalLayer using MetalSplatter (link below) - We had to make some optimizations within MetalSplatter's code too, to reduce the overhead of sorting points per render Then I added some finishing touches like the subtle rotation and parallax as they move around. I think it turned out pretty cool :) Overall, this took some effort, but we still got it done in less than a day. Hopefully your agent can follow in the footsteps of mine and do it much faster. Keep an eye out for the bigger writeup, which'll give your agent everything it needs. If you have any questions, drop em below!

Anshu

19,931 views • 1 month ago

Depth Any Video with Scalable Synthetic Data AI physicists and chemists continue to make strides in depth estimation from video. Check out this new paper featuring some impressive examples. See the thread for more details (unfortunately no code yet). Abstract: Video depth estimation has long been hindered by the scarcity of consistent and scalable ground truth data, leading to inconsistent and unreliable results. In this paper, we introduce Depth Any Video, a model that tackles the challenge through two key innovations. First, we develop a scalable synthetic data pipeline, capturing real-time video depth data from diverse game environments, yielding 40,000 video clips of 5-second duration, each with precise depth annotations. Second, we leverage the powerful priors of generative video diffusion models to handle real-world videos effectively, integrating advanced techniques such as rotary position encoding and flow matching to further enhance flexibility and efficiency. Unlike previous models, which are limited to fixed-length video sequences, our approach introduces a novel mixed-duration training strategy that handles videos of varying lengths and performs robustly across different frame rates 0 - even on single frames. At inference, we propose a depth interpolation method that enables our model to infer high-resolution video depth across sequences of up to 150 frames. Our model outperforms all previous generative depth models in terms of spatial accuracy and temporal consistency.

Depth Any Video with Scalable Synthetic Data AI physicists and chemists continue to make strides in depth estimation from video. Check out this new paper featuring some impressive examples. See the thread for more details (unfortunately no code yet). Abstract: Video depth estimation has long been hindered by the scarcity of consistent and scalable ground truth data, leading to inconsistent and unreliable results. In this paper, we introduce Depth Any Video, a model that tackles the challenge through two key innovations. First, we develop a scalable synthetic data pipeline, capturing real-time video depth data from diverse game environments, yielding 40,000 video clips of 5-second duration, each with precise depth annotations. Second, we leverage the powerful priors of generative video diffusion models to handle real-world videos effectively, integrating advanced techniques such as rotary position encoding and flow matching to further enhance flexibility and efficiency. Unlike previous models, which are limited to fixed-length video sequences, our approach introduces a novel mixed-duration training strategy that handles videos of varying lengths and performs robustly across different frame rates 0 - even on single frames. At inference, we propose a depth interpolation method that enables our model to infer high-resolution video depth across sequences of up to 150 frames. Our model outperforms all previous generative depth models in terms of spatial accuracy and temporal consistency.

MrNeRF

27,428 views • 1 year ago

🚀 The Segment Anything Model (SAM) has been upgraded to SAM2, featuring an efficient image encoder for segmenting images and videos. But does SAM2 outperform SAM1 in medical image and video segmentation? We're thrilled to present our paper "Segment Anything in Medical Images and Videos: Benchmark and Deployment"! We comprehensively benchmark SAM2 across 11 medical image modalities and videos. 📄 Paper: 💻 Code: **Highlights:** 1. SAM2 doesn’t always outperform SAM1 in 2D medical images, but excels in video segmentation, making it more accurate and efficient for 3D images, such as CT and MR scans. 2. MedSAM still outperforms SAM2 on most 2D modalities, but SAM2 surpasses MedSAM for 3D image segmentation in a slice-by-slice approach. 3. Segmentation performance varies with model size; sometimes the smallest model outperforms larger ones. 4. Fine-tuning SAM2 significantly boosts its performance for medical image segmentation. While SAM2 may struggle with challenging objects that have unclear boundaries or low contrast, it excels in generating good initial segmentation masks for common medical images and videos. However, the official interface doesn’t support medical data formats and has limitations on video length. To address this, we've developed a 3D Slicer Plugin and Gradio API for efficient 3D medical image and video segmentation. We invite you to try them out and provide feedback! 🔧 Deployment: - 3D Slicer Plugin: - Gradio API: (Note: Due to GPU limitations, the online API is available for only 12 hours and may be slow. We highly recommend deploying the Gradio API with your own computing resources: A big shoutout to Jun Ma (JunMa) who recently joined our UHN AI hub (UHN AI Hub) as Machine Learning Lead, and kudos to all co-authors: Sumin Kim, Feifei Li, Mohammed Baharoon (Mohammed Baharoon), Reza Asakereh, and Hongwei Lyu! This is true teamwork! Looking forward to collaborating with the community to advance 3D medical image and video segmentation foundation models! University Health Network U of T Department of Computer Science Department of Laboratory Medicine & Pathobiology Temerty Centre for AI in Medicine (T-CAIREM) Vector Institute #MedTech #AIinHealthcare #DeepLearning #MedicalImaging #SAM2 #MedSAM #AIResearch

🚀 The Segment Anything Model (SAM) has been upgraded to SAM2, featuring an efficient image encoder for segmenting images and videos. But does SAM2 outperform SAM1 in medical image and video segmentation? We're thrilled to present our paper "Segment Anything in Medical Images and Videos: Benchmark and Deployment"! We comprehensively benchmark SAM2 across 11 medical image modalities and videos. 📄 Paper: 💻 Code: Highlights: 1. SAM2 doesn’t always outperform SAM1 in 2D medical images, but excels in video segmentation, making it more accurate and efficient for 3D images, such as CT and MR scans. 2. MedSAM still outperforms SAM2 on most 2D modalities, but SAM2 surpasses MedSAM for 3D image segmentation in a slice-by-slice approach. 3. Segmentation performance varies with model size; sometimes the smallest model outperforms larger ones. 4. Fine-tuning SAM2 significantly boosts its performance for medical image segmentation. While SAM2 may struggle with challenging objects that have unclear boundaries or low contrast, it excels in generating good initial segmentation masks for common medical images and videos. However, the official interface doesn’t support medical data formats and has limitations on video length. To address this, we've developed a 3D Slicer Plugin and Gradio API for efficient 3D medical image and video segmentation. We invite you to try them out and provide feedback! 🔧 Deployment: - 3D Slicer Plugin: - Gradio API: (Note: Due to GPU limitations, the online API is available for only 12 hours and may be slow. We highly recommend deploying the Gradio API with your own computing resources: A big shoutout to Jun Ma (JunMa) who recently joined our UHN AI hub (UHN AI Hub) as Machine Learning Lead, and kudos to all co-authors: Sumin Kim, Feifei Li, Mohammed Baharoon (Mohammed Baharoon), Reza Asakereh, and Hongwei Lyu! This is true teamwork! Looking forward to collaborating with the community to advance 3D medical image and video segmentation foundation models! University Health Network U of T Department of Computer Science Department of Laboratory Medicine & Pathobiology Temerty Centre for AI in Medicine (T-CAIREM) Vector Institute #MedTech #AIinHealthcare #DeepLearning #MedicalImaging #SAM2 #MedSAM #AIResearch

Bo Wang

178,539 views • 2 years ago

This week is already so hot. 🔥 Massive release from Decart : Lucy 2.0 a World Editing Model running at 1080p, 30FPS in realtime. This is truly exciting, the era of real-time generative reality is here. We are moving from watching AI video to living inside AI video. A breakthrough model capable of transforming the visual world in real-time. Moving beyond offline rendering, Lucy 2.0 delivers high-fidelity 1080p video generation with near-zero latency. Lucy 2.0 literally "redraws" the entire world pixel-by-pixel, while you are watching it. e.g. If you want to be an anime character, it doesn't just put a mask on you. It turns your skin into anime skin, your hair into anime hair, and the lighting in your room into anime lighting. Lucy 2.0 is also trained to stop the generated video from slowly falling apart over time, so the same stream can run much longer without faces and details drifting. So why is this a "Massive Deal"? Traditional AI video-generation model takes a prompt, you wait 10–20 minutes, and the computer "bakes" a video for you. You couldn't touch it or change it while it was happening. But Lucy 2.0 works like a mirror. It happens in real-time (30 frames per second). There is no waiting. You move your hand, the AI character moves its hand instantly. The craziest part isn't the visuals; it's the physics. Usually, AI hallucinations are glitchy—hands merge into faces, walls melt. Lucy 2.0 understands how the world works without being told. It knows that if you take off a helmet, there is hair underneath. It knows that if you splash water, droplets fly. It learned "physics" just by watching millions of videos. The physical behavior you see emerges from learned visual dynamics, not from engineered geometry or explicit physics engines. Their official technical report explicitly states that the model does not use traditional 3D engines, depth maps, or wireframes. It is a "pure diffusion model."

This week is already so hot. 🔥 Massive release from Decart : Lucy 2.0 a World Editing Model running at 1080p, 30FPS in realtime. This is truly exciting, the era of real-time generative reality is here. We are moving from watching AI video to living inside AI video. A breakthrough model capable of transforming the visual world in real-time. Moving beyond offline rendering, Lucy 2.0 delivers high-fidelity 1080p video generation with near-zero latency. Lucy 2.0 literally "redraws" the entire world pixel-by-pixel, while you are watching it. e.g. If you want to be an anime character, it doesn't just put a mask on you. It turns your skin into anime skin, your hair into anime hair, and the lighting in your room into anime lighting. Lucy 2.0 is also trained to stop the generated video from slowly falling apart over time, so the same stream can run much longer without faces and details drifting. So why is this a "Massive Deal"? Traditional AI video-generation model takes a prompt, you wait 10–20 minutes, and the computer "bakes" a video for you. You couldn't touch it or change it while it was happening. But Lucy 2.0 works like a mirror. It happens in real-time (30 frames per second). There is no waiting. You move your hand, the AI character moves its hand instantly. The craziest part isn't the visuals; it's the physics. Usually, AI hallucinations are glitchy—hands merge into faces, walls melt. Lucy 2.0 understands how the world works without being told. It knows that if you take off a helmet, there is hair underneath. It knows that if you splash water, droplets fly. It learned "physics" just by watching millions of videos. The physical behavior you see emerges from learned visual dynamics, not from engineered geometry or explicit physics engines. Their official technical report explicitly states that the model does not use traditional 3D engines, depth maps, or wireframes. It is a "pure diffusion model."

Rohan Paul

12,761 views • 6 months ago

AI Is Moving Beyond “Generating Videos” — Toward “Generating Worlds” Over the past two years, AI video models have advanced at an astonishing pace. From Runway and Pika to Sora and Veo, AI-generated videos have become increasingly realistic and more consistent with the physical laws of the real world. Many people believe the next objective is simply to generate videos that are longer, sharper, and more lifelike. But if we take a step back, we can see that the real transformation is not happening in video itself. It is happening in world models. What Is a World Model? In 1943, psychologist Kenneth Craik proposed an idea that would influence artificial intelligence research for decades. He argued that the human brain does not merely react to the outside world. Instead, it maintains an internal model of how the world works. Because we have this internal model, we can predict the outcome of an action before we actually take it. Before crossing a road, we estimate whether a car will pass by. Before catching a ball, we predict its trajectory. These abilities come from continuously simulating the world in our minds, rather than relying entirely on trial and error. This idea later became known by a more formal term: World Model. A world model does not describe a single image or a fixed video clip. It is an internal representation capable of continuously simulating the rules and dynamics of the real world. Why Is AI Research Turning Toward World Models? Because predicting “what comes next” is becoming increasingly central to how AI systems work. Language models predict the next token. Image models predict the next step in the denoising process. Video models predict the next frame. A world model, however, attempts to predict something broader: What should the world look like in the next moment? In 2018, David Ha and Jürgen Schmidhuber proposed in their paper World Models that an intelligent agent could first learn a model of the world, and then use that internal model to plan its actions. The Dreamer series later demonstrated that many complex tasks could be learned by training agents inside an “imagined world.” At the same time, the development of video models such as Sora and Veo led researchers to another realization: A model capable of continuously generating video has already learned, at least implicitly, many of the rules governing the real world. As a result, these two research directions have gradually begun to converge. But Video Is Not Yet a World This is where the distinction is often misunderstood. For a world model to support meaningful real-time interaction, it must solve several critical problems. Most video models today are essentially answering one question: What should the next frame look like? A true world model needs to answer much more: What happens if I take one step forward? If I walk behind a building and then return, will the building still be there? If I suddenly change the camera angle, will the entire space remain consistent? If I enter a command such as: “Summon a dragon.” Will the world respond immediately? In other words, a world model must do more than generate content. It must understand space. It must understand time. It must understand causality. And it must understand interaction. Moving from watching to participating is where the real difficulty of world models begins. World Models Are Entering the Interactive Era One of the latest attempts in this direction is Alaya World, recently open-sourced by Alaya World, or Alaya Lab. Instead of generating a fixed video clip, it generates a world that users can explore in real time. Users can begin with text, an image, or a video, enter the generated scene, move freely through it, and introduce new prompts at any moment during generation. The world responds immediately. According to the publicly released information, Alaya World provides: Real-time streaming generation at 720p and 24 FPS Stable continuous exploration for more than one minute The ability to switch prompts and trigger skills or events during generation Model weights and inference code released under the Apache 2.0 License Training code and datasets planned for future release What makes these capabilities important is not simply the technical specifications. It is that the generated “world” can now support continuous interaction. The official demo shows that users can genuinely control, transform, and explore the generated environment. AI Is Evolving From a Tool Into an Environment Over the past few years, most discussions around AI have focused on content generation. Generating text. Generating images. Generating videos. But world models raise a fundamentally different question: Can AI generate an environment that people can inhabit, explore, and continuously evolve? If the answer is yes, the impact will extend far beyond video generation. Game development, robotics training, embodied intelligence, digital twins, virtual production, and many other fields could be transformed by the development of world models. World models are still at a very early stage. Yet from Craik’s proposal of an internal mental model more than eighty years ago to the emergence of today’s interactive world-generation systems, a clear evolutionary path is beginning to take shape. Perhaps what AI is ultimately learning has never been limited to images, videos, or language. Perhaps it is learning the world itself. References GitHub: Technical Report:

AI Is Moving Beyond “Generating Videos” — Toward “Generating Worlds” Over the past two years, AI video models have advanced at an astonishing pace. From Runway and Pika to Sora and Veo, AI-generated videos have become increasingly realistic and more consistent with the physical laws of the real world. Many people believe the next objective is simply to generate videos that are longer, sharper, and more lifelike. But if we take a step back, we can see that the real transformation is not happening in video itself. It is happening in world models. What Is a World Model? In 1943, psychologist Kenneth Craik proposed an idea that would influence artificial intelligence research for decades. He argued that the human brain does not merely react to the outside world. Instead, it maintains an internal model of how the world works. Because we have this internal model, we can predict the outcome of an action before we actually take it. Before crossing a road, we estimate whether a car will pass by. Before catching a ball, we predict its trajectory. These abilities come from continuously simulating the world in our minds, rather than relying entirely on trial and error. This idea later became known by a more formal term: World Model. A world model does not describe a single image or a fixed video clip. It is an internal representation capable of continuously simulating the rules and dynamics of the real world. Why Is AI Research Turning Toward World Models? Because predicting “what comes next” is becoming increasingly central to how AI systems work. Language models predict the next token. Image models predict the next step in the denoising process. Video models predict the next frame. A world model, however, attempts to predict something broader: What should the world look like in the next moment? In 2018, David Ha and Jürgen Schmidhuber proposed in their paper World Models that an intelligent agent could first learn a model of the world, and then use that internal model to plan its actions. The Dreamer series later demonstrated that many complex tasks could be learned by training agents inside an “imagined world.” At the same time, the development of video models such as Sora and Veo led researchers to another realization: A model capable of continuously generating video has already learned, at least implicitly, many of the rules governing the real world. As a result, these two research directions have gradually begun to converge. But Video Is Not Yet a World This is where the distinction is often misunderstood. For a world model to support meaningful real-time interaction, it must solve several critical problems. Most video models today are essentially answering one question: What should the next frame look like? A true world model needs to answer much more: What happens if I take one step forward? If I walk behind a building and then return, will the building still be there? If I suddenly change the camera angle, will the entire space remain consistent? If I enter a command such as: “Summon a dragon.” Will the world respond immediately? In other words, a world model must do more than generate content. It must understand space. It must understand time. It must understand causality. And it must understand interaction. Moving from watching to participating is where the real difficulty of world models begins. World Models Are Entering the Interactive Era One of the latest attempts in this direction is Alaya World, recently open-sourced by Alaya World, or Alaya Lab. Instead of generating a fixed video clip, it generates a world that users can explore in real time. Users can begin with text, an image, or a video, enter the generated scene, move freely through it, and introduce new prompts at any moment during generation. The world responds immediately. According to the publicly released information, Alaya World provides: Real-time streaming generation at 720p and 24 FPS Stable continuous exploration for more than one minute The ability to switch prompts and trigger skills or events during generation Model weights and inference code released under the Apache 2.0 License Training code and datasets planned for future release What makes these capabilities important is not simply the technical specifications. It is that the generated “world” can now support continuous interaction. The official demo shows that users can genuinely control, transform, and explore the generated environment. AI Is Evolving From a Tool Into an Environment Over the past few years, most discussions around AI have focused on content generation. Generating text. Generating images. Generating videos. But world models raise a fundamentally different question: Can AI generate an environment that people can inhabit, explore, and continuously evolve? If the answer is yes, the impact will extend far beyond video generation. Game development, robotics training, embodied intelligence, digital twins, virtual production, and many other fields could be transformed by the development of world models. World models are still at a very early stage. Yet from Craik’s proposal of an internal mental model more than eighty years ago to the emergence of today’s interactive world-generation systems, a clear evolutionary path is beginning to take shape. Perhaps what AI is ultimately learning has never been limited to images, videos, or language. Perhaps it is learning the world itself. References GitHub: Technical Report:

雪踏乌云

112,114 views • 13 days ago

STEVE-1: A Generative Model for Text-to-Behavior in Minecraft paper page: Constructing AI models that respond to text instructions is challenging, especially for sequential decision-making tasks. This work introduces an instruction-tuned Video Pretraining (VPT) model for Minecraft called STEVE-1, demonstrating that the unCLIP approach, utilized in DALL-E 2, is also effective for creating instruction-following sequential decision-making agents. STEVE-1 is trained in two steps: adapting the pretrained VPT model to follow commands in MineCLIP's latent space, then training a prior to predict latent codes from text. This allows us to finetune VPT through self-supervised behavioral cloning and hindsight relabeling, bypassing the need for costly human text annotations. By leveraging pretrained models like VPT and MineCLIP and employing best practices from text-conditioned image generation, STEVE-1 costs just $60 to train and can follow a wide range of short-horizon open-ended text and visual instructions in Minecraft. STEVE-1 sets a new bar for open-ended instruction following in Minecraft with low-level controls (mouse and keyboard) and raw pixel inputs, far outperforming previous baselines. We provide experimental evidence highlighting key factors for downstream performance, including pretraining, classifier-free guidance, and data scaling. All resources, including our model weights, training scripts, and evaluation tools are made available for further research.

STEVE-1: A Generative Model for Text-to-Behavior in Minecraft paper page: Constructing AI models that respond to text instructions is challenging, especially for sequential decision-making tasks. This work introduces an instruction-tuned Video Pretraining (VPT) model for Minecraft called STEVE-1, demonstrating that the unCLIP approach, utilized in DALL-E 2, is also effective for creating instruction-following sequential decision-making agents. STEVE-1 is trained in two steps: adapting the pretrained VPT model to follow commands in MineCLIP's latent space, then training a prior to predict latent codes from text. This allows us to finetune VPT through self-supervised behavioral cloning and hindsight relabeling, bypassing the need for costly human text annotations. By leveraging pretrained models like VPT and MineCLIP and employing best practices from text-conditioned image generation, STEVE-1 costs just $60 to train and can follow a wide range of short-horizon open-ended text and visual instructions in Minecraft. STEVE-1 sets a new bar for open-ended instruction following in Minecraft with low-level controls (mouse and keyboard) and raw pixel inputs, far outperforming previous baselines. We provide experimental evidence highlighting key factors for downstream performance, including pretraining, classifier-free guidance, and data scaling. All resources, including our model weights, training scripts, and evaluation tools are made available for further research.

AK

144,783 views • 3 years ago

Seedance 2.0 on FlovaAI =================== Prompt: [Reference Identity Lock] Image 1 is ONLY the main female protagonist. Her face, hairstyle, body type, and outfit must match Image 1 exactly and stay consistent for the entire video. Image 2 is ONLY a uniform reference. All four opponents wear the school uniform shown in Image 2. Never swap, merge, duplicate, or blend identities. The protagonist's identity comes ONLY from Image 1. The four opponents have NO reference images. They are defined by the text descriptions below. The four opponents must not resemble the protagonist, and they must not resemble each other. All five characters must remain clearly distinct and recognizable until the end. [Priority Order] 1. Preserve the protagonist's identity from Image 1. 2. Keep the four opponents visually distinct from her and from each other. 3. Maintain one continuous shot with no cuts. 4. Keep the classroom layout spatially consistent. 5. Make the action fast but readable and physically connected. 6. Keep the tone as a Korean school action drama, stylish but grounded. Korean school action drama classroom fight scene — 15 seconds, ONE CONTINUOUS SHOT, NO CUTS. A single uninterrupted handheld shot. No cuts, no scene transitions, no montage. The camera should feel handheld, with micro-jitters, slight rolling shutter, and raw unstable realism. The camera must physically travel through the same classroom space. Every transition must be motivated by camera movement, not editing. Whip pans are allowed, but they must not hide a cut. Do not teleport the camera or characters. The classroom layout and character positions must remain spatially consistent. Audio: No music. Only realistic school and classroom ambient sounds: old fluorescent light hum, distant hallway noise, ceiling fan, shoes scraping the floor, desks dragging, chair legs screeching, cloth friction, dull body impacts, and breathing that gradually becomes heavier. Breathing continues throughout the scene and keeps building. Lighting: Late afternoon in a Korean high school classroom. Mixed cool fluorescent light and warm sunlight through the windows. Dust floating in the sunlight. Soft fan shadows moving across desks and school uniforms. Main character: The Korean female high school student from Image 1, age 17–18. Cold, emotionless, calm, and intimidating. She barely speaks and does not scream during the fight. She remains composed from beginning to end. Her movements are efficient, explosive, and precise. Even if her frame is not large, she dominates through speed, timing, and accuracy. Main outfit: Exactly the outfit shown in Image 1. Do not change its colors, design, or details. Her jacket or outer layer is either removed and hanging on a chair, or worn in a slightly messy way. The action must be non-sexualized and combat-focused. Fabric movement, dust, sweat, wrinkles, and impact response should feel realistic. Opponent rules: Four Korean female high school students, all wearing the Hanlim Multi Art School uniform shown in Image 2. They have no reference images. Define them strictly by these descriptions and keep each one consistent: Opponent A: short black bob with straight bangs, medium build, round face. Opponent B: long straight hair tied in a high ponytail, tall and lean, sharp jawline. Opponent C: shoulder-length hair with side-swept bangs, slim build, narrow face. Opponent D: long wavy hair worn loose, slightly stocky and broad-shouldered. A, B, C, and D must each keep clearly different faces, hairstyles, body shapes, and silhouettes. They must not resemble the protagonist, and they must not resemble each other. No face duplication, no face merging, no identity confusion. Environment: An empty classroom at Hanlim Multi Art School, a Korean performing arts high school in Seoul. Green chalkboard, chalk tray, worn wooden desks, plastic chairs, classroom clock, class schedule poster, discipline/life-guidance posters, cleaning tools, blinds or curtains, wall study materials, and a slightly scuffed floor. Desks and chairs should react naturally to impacts, sliding, shaking, and collapsing when hit. Camera framing rules: Even during kicks, framing should stay around chest-level or eye-level. No low-angle shots under the skirt. Do not focus on legs, thighs, underwear, or fetish-like details. All action framing must prioritize faces, upper-body motion, impact, and spatial choreography. Continuous action and camera choreography: From 0 to 15 seconds, the fight continues without any cuts. The action should be stylish but readable, and every movement must be physically connected. 0–3s: The camera starts behind the protagonist at a slightly low handheld angle, drifting left through the classroom aisle. Opponent A grabs the protagonist's shoulder roughly and says in Korean: "야, 너 지금 뭐 하자는 거야?" The protagonist silently turns and lands one hard straight punch to A's face. At impact, use a very brief 15% slow motion: cheek ripple, dust particles, deep thud. A falls sideways into a desk. The camera dips slightly from the shock, then whip-pans right without cutting. 3–6s: Opponent B charges in from the right. The protagonist steps forward instead of retreating. A short body shot to the stomach. Immediate uppercut to the chin. Without pausing, she drives forward into a flying knee to B's chest. B is thrown backward across or into a desk. The camera follows the forward motion low, then rebounds upward with the impact. 6–9s: Opponent D attacks with two fast punches. The protagonist deflects both strikes with her arms, then flows into a turning backfist to D's face. As D staggers, she continues the same rotation into a spinning back elbow that lands hard on D's jaw or temple. D crashes sideways into two or three desks. The camera arcs around her shoulder and jitters slightly at each impact. No cuts. 9–12s: Opponent C rushes in from the chalkboard side. The protagonist clearly grabs C's collar with her left hand. C's face must be fully visible from the front and clearly different from the protagonist. The protagonist lands one short, hard punch to C's face, then immediately throws a powerful high kick or flying high kick into C's chest. The force sends C backward into the green chalkboard. The protagonist remains in the foreground and never touches the board. The protagonist's face should be side-profile or partially obscured. C's face should be clearly visible from the front at the moment of impact. Their faces must never overlap in frame. Use a very brief 20% slow motion at the chalkboard impact: chalk dust bursts outward, and C slides down the board. The camera pushes up with the impact, then tilts down as C slides. 12–15s: Through the chalk dust, the camera hard-pans right. D makes one final charge. The protagonist sidesteps and lands a tight uppercut to D's chin, followed immediately by a cross. D crashes into a row of desks, causing a chain reaction of collapsing desks and chairs. The camera drifts forward slowly. The protagonist adjusts her loose tie or ribbon and brushes chalk dust off her shoulder. Her expression stays cold and serious. She walks past the camera and exits the frame. Dust floats in the sunlight. Natural ending. =================== Made with Flova #FlovaAI #FlovaCPP

TSUBAKI

17,869 views • 12 days ago

This Kling 3.0 UGC workflow is absolutely insane 🤯 It generates hyper-realistic UGC-style ad videos from a single text prompt — talking heads, unboxings, testimonials, tutorials — in one shot. Perfect for DTC brands and agencies who need to scale UGC without paying $500 per UGC video. Here's the problem: You're either paying creators who take weeks to deliver, or you're stitching together janky 5-second AI clips that look obviously fake. Kling 3.0 solves it: → Write a prompt with character, scene, and dialogue → Get a 10-15 second UGC video with realistic facial expressions → Native voice control — tone, pacing, emotion per line → Multi-shot scenes in one generation (hook → demo → CTA) → Paste negative rules to kill AI artifacts instantly No creator fees. No stitching clips. No uncanny valley. What you get: → 6 copy-paste prompt templates for every UGC format →Voice & dialogue cheat sheet with tone keywords → Multi-shot ad structure (hook → problem → solution → CTA) → Image-to-video workflow for branded product shots → A "where it fails" section so you don't waste credits → ChatGPT template to convert any rough idea into a Kling prompt I built a full prompting guide for UGC ads on TikTok and Meta. Want the guide? > Like this post > Comment "KLING" And I'll send it over (must be following so I can DM)

This Kling 3.0 UGC workflow is absolutely insane 🤯 It generates hyper-realistic UGC-style ad videos from a single text prompt — talking heads, unboxings, testimonials, tutorials — in one shot. Perfect for DTC brands and agencies who need to scale UGC without paying $500 per UGC video. Here's the problem: You're either paying creators who take weeks to deliver, or you're stitching together janky 5-second AI clips that look obviously fake. Kling 3.0 solves it: → Write a prompt with character, scene, and dialogue → Get a 10-15 second UGC video with realistic facial expressions → Native voice control — tone, pacing, emotion per line → Multi-shot scenes in one generation (hook → demo → CTA) → Paste negative rules to kill AI artifacts instantly No creator fees. No stitching clips. No uncanny valley. What you get: → 6 copy-paste prompt templates for every UGC format →Voice & dialogue cheat sheet with tone keywords → Multi-shot ad structure (hook → problem → solution → CTA) → Image-to-video workflow for branded product shots → A "where it fails" section so you don't waste credits → ChatGPT template to convert any rough idea into a Kling prompt I built a full prompting guide for UGC ads on TikTok and Meta. Want the guide? > Like this post > Comment "KLING" And I'll send it over (must be following so I can DM)

Mike Futia

36,105 views • 5 months ago

Before the week ends, let's acknowledge one of the most INSANE week ever for open AI, with 25+ notable open-weight drops across every modality: 🧠 LLMs → NVIDIA Nemotron 3 Ultra: 550B hybrid Mamba-MoE, only 55B active, 1M context, MMLU 89.1. NVFP4 variant claims ~5x throughput on Blackwell. First openly-weighted 550B hybrid Mamba-Transformer, closing the gap with frontier closed models. → Google Gemma 4 12B: fully open dense any-to-any (text/image/audio/video), 256k context, encoder-free, 140+ languages, AIME 2026 at 77.5. Shipped with a 23-checkpoint QAT wave (mobile ONNX + MLX). Most deployable model of the week. → StepFun Step-3.7-Flash: 198B sparse MoE VLM, ~11B active, SWE-Bench PRO 56.3. Apache 2.0. → Liquid AI LFM2.5-8B-A1B: edge MoE, just 1.5B active, 128k ctx, MATH500 88.8, MLX-ready. Best on-device option this week. → JetBrains Mellum2-12B-A2.5B-Thinking: their first open MoE, near-Qwen3-14B coding at 2.5B active. Apache 2.0. 🎨 Image gen (the surprise of the week) → Ideogram 4: their FIRST-EVER open weights. 9.3B flow-matching DiT trained from scratch. #2 overall behind GPT Image 2, top open-weight model on Design Arena + LMArena. Strongest open checkpoint for text-rich images, full stop. It has taste. Still can't believe this is open weights. 🔊 Audio & Speech (a breakout week for open TTS, 4 labs shipped) → Boson Higgs Audio v3 4B: 102 languages, 21 emotions, singing/whispering/shouting, sub-second TTFA. → RedNote dots.tts: the only fully continuous (no codec) open TTS pipeline, Apache 2.0. → Google Magenta RealTime 2: real-time music gen, <200ms latency, text+audio+MIDI. multimodalart ported it to PyTorch within hours with live ZeroGPU demos. → NVIDIA Nemotron-3.5 ASR: 600M streaming, 17x more concurrent streams vs Parakeet RNNT 1.1B. 👁️ Vision & VLMs → PaddleOCR-VL-1.6: SOTA document parsing at 1B params, Apache 2.0. → Baidu NAVA: 6.3B joint audio-video gen, best-in-class A/V sync, Apache 2.0. 🎬 Video, 3D & World Models → NVIDIA Cosmos3-Super: 64B omnimodal world model coupling action trajectories with video+audio gen, for Physical AI. → JD JoyAI-Echo: up to 5-min multi-shot text-to-video on LTX-2.3. → ByteDance Bernini-R + VAST TripoSplat (single-image-to-3D Gaussian splats, MIT).

Before the week ends, let's acknowledge one of the most INSANE week ever for open AI, with 25+ notable open-weight drops across every modality: 🧠 LLMs → NVIDIA Nemotron 3 Ultra: 550B hybrid Mamba-MoE, only 55B active, 1M context, MMLU 89.1. NVFP4 variant claims ~5x throughput on Blackwell. First openly-weighted 550B hybrid Mamba-Transformer, closing the gap with frontier closed models. → Google Gemma 4 12B: fully open dense any-to-any (text/image/audio/video), 256k context, encoder-free, 140+ languages, AIME 2026 at 77.5. Shipped with a 23-checkpoint QAT wave (mobile ONNX + MLX). Most deployable model of the week. → StepFun Step-3.7-Flash: 198B sparse MoE VLM, ~11B active, SWE-Bench PRO 56.3. Apache 2.0. → Liquid AI LFM2.5-8B-A1B: edge MoE, just 1.5B active, 128k ctx, MATH500 88.8, MLX-ready. Best on-device option this week. → JetBrains Mellum2-12B-A2.5B-Thinking: their first open MoE, near-Qwen3-14B coding at 2.5B active. Apache 2.0. 🎨 Image gen (the surprise of the week) → Ideogram 4: their FIRST-EVER open weights. 9.3B flow-matching DiT trained from scratch. #2 overall behind GPT Image 2, top open-weight model on Design Arena + LMArena. Strongest open checkpoint for text-rich images, full stop. It has taste. Still can't believe this is open weights. 🔊 Audio & Speech (a breakout week for open TTS, 4 labs shipped) → Boson Higgs Audio v3 4B: 102 languages, 21 emotions, singing/whispering/shouting, sub-second TTFA. → RedNote dots.tts: the only fully continuous (no codec) open TTS pipeline, Apache 2.0. → Google Magenta RealTime 2: real-time music gen, <200ms latency, text+audio+MIDI. multimodalart ported it to PyTorch within hours with live ZeroGPU demos. → NVIDIA Nemotron-3.5 ASR: 600M streaming, 17x more concurrent streams vs Parakeet RNNT 1.1B. 👁️ Vision & VLMs → PaddleOCR-VL-1.6: SOTA document parsing at 1B params, Apache 2.0. → Baidu NAVA: 6.3B joint audio-video gen, best-in-class A/V sync, Apache 2.0. 🎬 Video, 3D & World Models → NVIDIA Cosmos3-Super: 64B omnimodal world model coupling action trajectories with video+audio gen, for Physical AI. → JD JoyAI-Echo: up to 5-min multi-shot text-to-video on LTX-2.3. → ByteDance Bernini-R + VAST TripoSplat (single-image-to-3D Gaussian splats, MIT).

Victor M

540,078 views • 1 month ago

Not escaping the city—owning every street. 🚘 Generated this masterclass cinematic visuals with BudgetPixel AI Prompt 👇 GTA 6 Style: Neon Heist Escape Style: Ultra-realistic open-world video game cinematic, next-gen graphics, vibrant neon city, cinematic lighting, ray tracing, dynamic reflections, realistic physics, motion blur, dramatic camera work, 4K HDR, 60 FPS, intense sound design, no text, no watermark. 0.0–2.0s | Scene 1 – Neon Arrival A customized supercar drifts into a neon-lit downtown street at night. Low-angle tracking shot with glowing reflections on wet roads. The confident main character steps out wearing a modern street outfit and sunglasses. 2.0–4.0s | Scene 2 – The Heist Inside a luxurious casino, the character grabs a mysterious glowing briefcase while alarms suddenly activate. Fast cinematic close-ups, flashing red emergency lights, NPCs running in panic. 4.0–6.0s | Scene 3 – High-Speed Escape The character jumps into the supercar and launches into a high-speed chase through palm-lined boulevards. Dynamic drone shots, tire smoke, sharp drifts, police cars closing in. 6.0–8.5s | Scene 4 – Rooftop Action The chase transitions to rooftops where the character performs a cinematic parkour leap between skyscrapers at sunset. Slow motion during the jump, dramatic city skyline in the background. 8.5–11.0s | Scene 5 – Beach Pursuit A high-performance sports bike races along a crowded tropical beach promenade. Helicopters search overhead while the camera alternates between FPV and wide cinematic angles. 11.0–13.0s | Scene 6 – Epic Showdown The character stands on a luxury yacht in the marina as rivals arrive by speedboats. Intense cinematic standoff with realistic water physics, golden-hour lighting, and wind-blown clothing. 13.0–15.0s | Scene 7 – Legendary Ending The hero drives toward the glowing city skyline as fireworks illuminate the night. The camera pulls back into a breathtaking aerial shot, ending with a powerful cinematic freeze-frame and an epic orchestral finish.This prompt is designed to capture the feel of a modern open-world crime game without copying any copyrighted characters, logos, or specific story elements.

Not escaping the city—owning every street. 🚘 Generated this masterclass cinematic visuals with BudgetPixel AI Prompt 👇 GTA 6 Style: Neon Heist Escape Style: Ultra-realistic open-world video game cinematic, next-gen graphics, vibrant neon city, cinematic lighting, ray tracing, dynamic reflections, realistic physics, motion blur, dramatic camera work, 4K HDR, 60 FPS, intense sound design, no text, no watermark. 0.0–2.0s | Scene 1 – Neon Arrival A customized supercar drifts into a neon-lit downtown street at night. Low-angle tracking shot with glowing reflections on wet roads. The confident main character steps out wearing a modern street outfit and sunglasses. 2.0–4.0s | Scene 2 – The Heist Inside a luxurious casino, the character grabs a mysterious glowing briefcase while alarms suddenly activate. Fast cinematic close-ups, flashing red emergency lights, NPCs running in panic. 4.0–6.0s | Scene 3 – High-Speed Escape The character jumps into the supercar and launches into a high-speed chase through palm-lined boulevards. Dynamic drone shots, tire smoke, sharp drifts, police cars closing in. 6.0–8.5s | Scene 4 – Rooftop Action The chase transitions to rooftops where the character performs a cinematic parkour leap between skyscrapers at sunset. Slow motion during the jump, dramatic city skyline in the background. 8.5–11.0s | Scene 5 – Beach Pursuit A high-performance sports bike races along a crowded tropical beach promenade. Helicopters search overhead while the camera alternates between FPV and wide cinematic angles. 11.0–13.0s | Scene 6 – Epic Showdown The character stands on a luxury yacht in the marina as rivals arrive by speedboats. Intense cinematic standoff with realistic water physics, golden-hour lighting, and wind-blown clothing. 13.0–15.0s | Scene 7 – Legendary Ending The hero drives toward the glowing city skyline as fireworks illuminate the night. The camera pulls back into a breathtaking aerial shot, ending with a powerful cinematic freeze-frame and an epic orchestral finish.This prompt is designed to capture the feel of a modern open-world crime game without copying any copyrighted characters, logos, or specific story elements.

Ai Girllie

27,172 views • 18 days ago

This guy built a visual scanner that reads 468 points on his face and 42 points on his hands from a regular webcam and turns them into a cloud of thousands of particles right between his palms. Inside, MediaPipe and TouchDesigner are linked: the first captures hands and face from the webcam with high accuracy, the second turns those coordinates into a live plane and feeds it into a POP system that instantly generates a swarm of particles in the shape of a head. No studio, no render farmer, no VR headset. Just a laptop, a webcam, and 1 TouchDesigner session. And traditional VJ studios keep teams of 5 people on a setup with lighting, custom hardware, and commercial plugins, while his expenses are only a TouchDesigner subscription and a regular USB camera. One laptop runs MediaPipe and TouchDesigner simultaneously, holds the camera stream at 60 FPS without drops, and in parallel processes 468 face points + 21 points on each hand. The camera captures frame after frame, MediaPipe in real time sends TouchDesigner the finger coordinates and face geometry, and the POP operator inside the engine translates those numbers into thousands of particle points with colors from bright pink to gold. This setup immediately defines the role of the tool and the limits of its autonomy. It knows where the fingertips are at every moment of the frame. It knows how to read the face geometry at any angle to the camera. It knows how to draw a swarm of particles between them with the right color and contour. → MediaPipe pulls 468 points from the face and 21 points from each hand, 60 times per second → TouchDesigner receives those coordinates, builds a virtual rectangle between the fingertips, and feeds it into the POP system → POP generates thousands of particle points in the shape of a head, coloring them in a gradient from bright pink to gold → The HUD layer adds green corners and a blue neon frame, styling the image like an AR interface → All layers assemble into 1 real-time frame that projects back onto the video in the camera window → The final image is recorded to a file or broadcast to a projector for a live installation And only when the guy spreads his hands wider does the plane between the palms stretch; brings them together, it narrows. Otherwise the system runs on its own. And when he moves from his home room to a concert hall, the same laptop with the same webcam launches the same TouchDesigner session in just 5 minutes, without reconfiguration, without a new team, and without a single line of new code. In his work setup there is no studio of his own and no team for assembly. On the desk sits a laptop with a webcam, on top run MediaPipe and TouchDesigner with POP operators, and the same setup through a USB camera moves to any concert without a new configuration. Out of everything I have seen this year, this is the cleanest Creative Coding setup on 1 laptop: 0 render farms, 0 studio lighting, and between them 3 libraries, thousands of particle points, and 1 webcam.

This guy built a visual scanner that reads 468 points on his face and 42 points on his hands from a regular webcam and turns them into a cloud of thousands of particles right between his palms. Inside, MediaPipe and TouchDesigner are linked: the first captures hands and face from the webcam with high accuracy, the second turns those coordinates into a live plane and feeds it into a POP system that instantly generates a swarm of particles in the shape of a head. No studio, no render farmer, no VR headset. Just a laptop, a webcam, and 1 TouchDesigner session. And traditional VJ studios keep teams of 5 people on a setup with lighting, custom hardware, and commercial plugins, while his expenses are only a TouchDesigner subscription and a regular USB camera. One laptop runs MediaPipe and TouchDesigner simultaneously, holds the camera stream at 60 FPS without drops, and in parallel processes 468 face points + 21 points on each hand. The camera captures frame after frame, MediaPipe in real time sends TouchDesigner the finger coordinates and face geometry, and the POP operator inside the engine translates those numbers into thousands of particle points with colors from bright pink to gold. This setup immediately defines the role of the tool and the limits of its autonomy. It knows where the fingertips are at every moment of the frame. It knows how to read the face geometry at any angle to the camera. It knows how to draw a swarm of particles between them with the right color and contour. → MediaPipe pulls 468 points from the face and 21 points from each hand, 60 times per second → TouchDesigner receives those coordinates, builds a virtual rectangle between the fingertips, and feeds it into the POP system → POP generates thousands of particle points in the shape of a head, coloring them in a gradient from bright pink to gold → The HUD layer adds green corners and a blue neon frame, styling the image like an AR interface → All layers assemble into 1 real-time frame that projects back onto the video in the camera window → The final image is recorded to a file or broadcast to a projector for a live installation And only when the guy spreads his hands wider does the plane between the palms stretch; brings them together, it narrows. Otherwise the system runs on its own. And when he moves from his home room to a concert hall, the same laptop with the same webcam launches the same TouchDesigner session in just 5 minutes, without reconfiguration, without a new team, and without a single line of new code. In his work setup there is no studio of his own and no team for assembly. On the desk sits a laptop with a webcam, on top run MediaPipe and TouchDesigner with POP operators, and the same setup through a USB camera moves to any concert without a new configuration. Out of everything I have seen this year, this is the cleanest Creative Coding setup on 1 laptop: 0 render farms, 0 studio lighting, and between them 3 libraries, thousands of particle points, and 1 webcam.

Blaze

38,242 views • 2 months ago