正在加载视频...

视频加载失败

返回首页

Thinking with Camera A Unified Multimodal Model for Camera-Centric Understanding and Generation

_akhaliq's profile picture

AK

529,114 subscribers

39,397 次观看 • 11 个月前 •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

3 条评论

AK 的头像
AK11 个月前

discuss with author:

Chen Change Loy 的头像
Chen Change Loy11 个月前

Thanks @_akhaliq for sharing! Paper: Project Page: Code:

Min Chon Chi 的头像
Min Chon Chi11 个月前

Interesting topic, but the media failed to load.

相关视频

ReViV: Reconstructing the Viewer and the View in 4D from Monocular Egocentric Video(ECCV2026) ReViV reconstructs viewer-centric human motion (body, hand, and gaze) and view-centric scene geometry (camera and depth) from a single egocentric RGB video in a unified feed-forward model. For visualization, the initial camera pose and camera intrinsics are estimated with GeoCalib.
0:54

ReViV: Reconstructing the Viewer and the View in 4D from Monocular Egocentric Video(ECCV2026) ReViV reconstructs viewer-centric human motion (body, hand, and gaze) and view-centric scene geometry (camera and depth) from a single egocentric RGB video in a unified feed-forward model. For visualization, the initial camera pose and camera intrinsics are estimated with GeoCalib.

Ryohei Sasaki@engineer

15,096 次观看 • 1 个月前

Let's remove VAEs and ViTs from video models! 🚀 Introducing 𝗣𝗶𝘅𝗲𝗹𝗨𝗠𝗠: an encoder-free unified multimodal model for image and video understanding and generation, directly in pixel space. Code and model available today! 🌐 📄
0:28

Let's remove VAEs and ViTs from video models! 🚀 Introducing 𝗣𝗶𝘅𝗲𝗹𝗨𝗠𝗠: an encoder-free unified multimodal model for image and video understanding and generation, directly in pixel space. Code and model available today! 🌐 📄

Cong Wei

71,377 次观看 • 1 天前

Introducing Omni, one unified model can support any-to-any multimodal modeling, including multimodal understanding, image/video generation and editing, world modeling and 3D reconstruction. All in one that adopts standard mixture-of-experts arch with only 3B activations.
1:20

Introducing Omni, one unified model can support any-to-any multimodal modeling, including multimodal understanding, image/video generation and editing, world modeling and 3D reconstruction. All in one that adopts standard mixture-of-experts arch with only 3B activations.

Ceyuan Yang

32,808 次观看 • 5 个月前

We present MMaDA, first diffusion that unifies text reasoning, multimodal understanding, and image generation through Mixed Long-CoT, and unified RL - UniGRPO. 📚 Paper: 💻 Code: 📦 Model:
0:25

We present MMaDA, first diffusion that unifies text reasoning, multimodal understanding, and image generation through Mixed Long-CoT, and unified RL - UniGRPO. 📚 Paper: 💻 Code: 📦 Model:

Ling Yang

87,719 次观看 • 1 年前

Woah. First person camera view being warped to a third person camera view using a unified flow and matching model.
1:29

Woah. First person camera view being warped to a third person camera view using a unified flow and matching model.

Bilawal Sidhu

339,712 次观看 • 1 年前

(1/10) 🔥Thrilled to introduce OneDiffusion—our latest work in unified diffusion modeling! 🚀 This model bridges the gap between image synthesis and understanding, excelling in a wide range of tasks: T2I, conditional generation, image understanding, identity preservation, multiview generation, and even camera pose estimation. Learn more at: Project: arXiv: Code (on the way):
1:22

(1/10) 🔥Thrilled to introduce OneDiffusion—our latest work in unified diffusion modeling! 🚀 This model bridges the gap between image synthesis and understanding, excelling in a wide range of tasks: T2I, conditional generation, image understanding, identity preservation, multiview generation, and even camera pose estimation. Learn more at: Project: arXiv: Code (on the way):

Jiasen Lu

33,426 次观看 • 1 年前

Introducing Atlas: The world's first multimodal world model that generates image and video frames with pixel-perfect camera control and reconstructs them in 3D. Model the world, move the camera, and simulate space & time.
1:25

Introducing Atlas: The world's first multimodal world model that generates image and video frames with pixel-perfect camera control and reconstructs them in 3D. Model the world, move the camera, and simulate space & time.

World Labs

10,827,951 次观看 • 1 个月前

One image + text + camera trajectory = controllable worlds. All on a single GPU. Our research team just released SANA-WM, a 2.6B open source world model natively trained for 60-second video generation with precise camera control.
0:20

One image + text + camera trajectory = controllable worlds. All on a single GPU. Our research team just released SANA-WM, a 2.6B open source world model natively trained for 60-second video generation with precise camera control.

NVIDIA AI

93,282 次观看 • 4 个月前

Big ideas deserve a powerful model. MiniMax H3 is now available on Fotor. From cinematic brand films and AI storytelling to ecommerce marketing, animation, and game-inspired visuals, MiniMax H3 is built to handle every stage of your creative workflow. With native multimodal understanding, multi-reference generation, advanced character and camera control, and expressive voice capabilities, turning complex ideas into production-ready content has never been easier. One model. Endless creative possibilities. Try MiniMax H3 on Fotor today.
0:29

Big ideas deserve a powerful model. MiniMax H3 is now available on Fotor. From cinematic brand films and AI storytelling to ecommerce marketing, animation, and game-inspired visuals, MiniMax H3 is built to handle every stage of your creative workflow. With native multimodal understanding, multi-reference generation, advanced character and camera control, and expressive voice capabilities, turning complex ideas into production-ready content has never been easier. One model. Endless creative possibilities. Try MiniMax H3 on Fotor today.

Fotor

81,660 次观看 • 2 个月前

SpatialTrackerV2: unified, end-to-end 3D point tracking model which simultaneously estimates Camera Motion, Consistent Geometry and Pixel-wise 3D Trajectories.
0:35

SpatialTrackerV2: unified, end-to-end 3D point tracking model which simultaneously estimates Camera Motion, Consistent Geometry and Pixel-wise 3D Trajectories.

Bilawal Sidhu

20,346 次观看 • 1 年前

"YoNoSplat: You Only Need One Model for Feedforward 3D Gaussian Splatting" TL;DR: a unified 3D Gaussian splatting model that reconstructs high-quality scene geometry and camera poses from unposed/uncalibrated images in a single forward pass.
0:38

"YoNoSplat: You Only Need One Model for Feedforward 3D Gaussian Splatting" TL;DR: a unified 3D Gaussian splatting model that reconstructs high-quality scene geometry and camera poses from unposed/uncalibrated images in a single forward pass.

Alexandre Morgand

15,035 次观看 • 7 个月前

Soooo my wonderful roommate @hypnojockCJ took me under last night... on camera. We're thinking of starting a hypno-centric JFF. 👀👀 So... Stay tuned!
0:57

Sensitive content

Soooo my wonderful roommate @hypnojockCJ took me under last night... on camera. We're thinking of starting a hypno-centric JFF. 👀👀 So... Stay tuned!

Leather Canuck™ TeddyHawk

20,570 次观看 • 2 年前

DINO-X: A Unified Vision Model for Open-World Object Detection and Understanding TL;DR: DINO-X Pro: sota model with enhanced perception capabilities for various scenarios; DINO-X Edge: model optimized for faster inference speed and better suited for deployment on edge devices
0:45

DINO-X: A Unified Vision Model for Open-World Object Detection and Understanding TL;DR: DINO-X Pro: sota model with enhanced perception capabilities for various scenarios; DINO-X Edge: model optimized for faster inference speed and better suited for deployment on edge devices

Alexandre Morgand

60,763 次观看 • 1 年前

BulletTime Decoupled Control of Time and Camera Pose for Video Generation
1:36

BulletTime Decoupled Control of Time and Camera Pose for Video Generation

AK

11,745 次观看 • 10 个月前

Meta presents TUNA Taming Unified Visual Representations for Native Unified Multimodal Models
1:11

Meta presents TUNA Taming Unified Visual Representations for Native Unified Multimodal Models

AK

32,796 次观看 • 10 个月前

🚀UniWorld: a unified model that skips VAEs and uses semantic features from SigLIP! Using just 1% of BAGEL’s data, it outperforms on image editing and excels in understanding & generation. 🌟Now data, model, training & evaluation script are open-source!
2:06

🚀UniWorld: a unified model that skips VAEs and uses semantic features from SigLIP! Using just 1% of BAGEL’s data, it outperforms on image editing and excels in understanding & generation. 🌟Now data, model, training & evaluation script are open-source!

Bin Lin

22,345 次观看 • 1 年前

🚀 We just released LTX-Video 13B – our most powerful open-source video generation model yet. – 13B parameters – Multiscale rendering for better fine details – Better motion and scene understanding – Still fast - runs on local GPUs – Supports keyframes, camera/character motion & multi-shot sequences
2:20

🚀 We just released LTX-Video 13B – our most powerful open-source video generation model yet. – 13B parameters – Multiscale rendering for better fine details – Better motion and scene understanding – Still fast - runs on local GPUs – Supports keyframes, camera/character motion & multi-shot sequences

Yoav HaCohen

182,962 次观看 • 1 年前