正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

🚨 Just built a system that reconstructs 4D NBA plays from a single broadcast video. 📦 Pipeline includes: • Object Detection • Camera Stabilization • Pose Tracking • Homography • Ball Path See more examples on my page—happy to answer any questions below.

Effxcts

1,645 subscribers

10,378 次观看 • 1 年前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

6 条评论

Daniel 的头像

Daniel1 年前

lit 🔥

Finn 的头像

Finn1 年前

I was playing around a bit with YOLO, how do you get from 2d poses to 3d, so lift them? Havent really figured that out. Any repo you use?

Effxcts 的头像

Effxcts1 年前

There are a bunch out there. Best one for video right now is probably GVHMR. There’s also Vid2Avatar and others that are coming out from CVPR 2025

Finn 的头像

Finn1 年前

and how do you deal with the relation of the players on the field? so their position. Been playing around with DPVO.

Effxcts 的头像

Effxcts1 年前

Camera stabilization and then homography. Couldn’t get DVPO installed on windows so did it that way

BIG P, THE TIGER 🐅 的头像

BIG P, THE TIGER 🐅1 年前

holy cow this is awesome

相关视频

A video is compressed 4D data. 🌍 Today, we’re sharing a preview of our latest research: 4D reconstruction from a single monocular video. 🎬 From one casually captured video, our system reconstructs a dynamic, freely viewable 4D Gaussian scene—recovering geometry, appearance, camera motion, and object motion across space and time. No multi-camera rig. Just video → 4DGS. The world has already been captured in video. The next step is to unlock the 4D structure inside it. Watch the result below. 👇 #4DGS #GaussianSplatting #CVPR #CVPR2026 #AIGC #AI #3DGS

A video is compressed 4D data. 🌍 Today, we’re sharing a preview of our latest research: 4D reconstruction from a single monocular video. 🎬 From one casually captured video, our system reconstructs a dynamic, freely viewable 4D Gaussian scene—recovering geometry, appearance, camera motion, and object motion across space and time. No multi-camera rig. Just video → 4DGS. The world has already been captured in video. The next step is to unlock the 4D structure inside it. Watch the result below. 👇 #4DGS #GaussianSplatting #CVPR #CVPR2026 #AIGC #AI #3DGS

InSpatio｜影溯

64,926 次观看 • 3 个月前

Introducing HO-Cap: A Capture System and Dataset for 3D Reconstruction and Pose Tracking of Hand-Object Interaction. We built a multi-camera system and a semi-automatic method for annotating the shape and pose of hands and objects Project page:

Introducing HO-Cap: A Capture System and Dataset for 3D Reconstruction and Pose Tracking of Hand-Object Interaction. We built a multi-camera system and a semi-automatic method for annotating the shape and pose of hands and objects Project page:

Yu Xiang

57,604 次观看 • 1 年前

Reshoot any video in 4D. Convert any video into a 4D Gaussian Splat. Create a new camera path in Comfy. Get the same moment back from the new camera. Four ComfyUI nodes, local, open weights, built and run end-to-end on a 3090. #OpenSourceHiggsfield 🧵

Reshoot any video in 4D. Convert any video into a 4D Gaussian Splat. Create a new camera path in Comfy. Get the same moment back from the new camera. Four ComfyUI nodes, local, open weights, built and run end-to-end on a 3090. #OpenSourceHiggsfield 🧵

metamike

15,397 次观看 • 4 天前

🚨 Someone just built a fully open-source mocap system that works with any camera. It's called FreeMoCap, a markerless 3D tracking system that runs on ordinary webcams. It turns multiple camera feeds into research-grade skeletal data automatically. 100% Open Source.

🚨 Someone just built a fully open-source mocap system that works with any camera. It's called FreeMoCap, a markerless 3D tracking system that runs on ordinary webcams. It turns multiple camera feeds into research-grade skeletal data automatically. 100% Open Source.

Simplifying AI

363,000 次观看 • 7 个月前

AI BASKETBALL ANALYSIS. A FULL COMPUTER VISION SYSTEM. BUILT ON YOLO, OPENCV, AND PYTHON. Take any NBA broadcast. Any camera angle. Any resolution. Feed it into the system. YOLO finds every player and the ball. Frame by frame. No manual annotation. No pre-labeled data.The model just sees the court and understands it. Zero-shot classification looks at jersey pixels and decides who plays for whom. Two teams separated in milliseconds. Without ever being told the team names. But here's where it gets wild. Court keypoint detection identifies the geometry of the playing surface. Homography transforms the broadcast camera into a top-down tactical map. Real meters. Real coordinates. From a flat 2D video. Now every player has a position on a real court. Every movement becomes a data point. Every pass becomes a vector. Speed. Distance. Possession time. Pass networks. All extracted from nothing but pixels. No GPS. No chips in the ball. No million-dollar Second Spectrum setup. A Python script and a GPU. The full pipeline is open source. The tutorial walks through every line of code. From detection to transformation to analytics output. Teams used to pay six figures for this data. Now you build it in an afternoon. Computer vision didn't just enter basketball. It democratized it.

AI BASKETBALL ANALYSIS. A FULL COMPUTER VISION SYSTEM. BUILT ON YOLO, OPENCV, AND PYTHON. Take any NBA broadcast. Any camera angle. Any resolution. Feed it into the system. YOLO finds every player and the ball. Frame by frame. No manual annotation. No pre-labeled data.The model just sees the court and understands it. Zero-shot classification looks at jersey pixels and decides who plays for whom. Two teams separated in milliseconds. Without ever being told the team names. But here's where it gets wild. Court keypoint detection identifies the geometry of the playing surface. Homography transforms the broadcast camera into a top-down tactical map. Real meters. Real coordinates. From a flat 2D video. Now every player has a position on a real court. Every movement becomes a data point. Every pass becomes a vector. Speed. Distance. Possession time. Pass networks. All extracted from nothing but pixels. No GPS. No chips in the ball. No million-dollar Second Spectrum setup. A Python script and a GPU. The full pipeline is open source. The tutorial walks through every line of code. From detection to transformation to analytics output. Teams used to pay six figures for this data. Now you build it in an afternoon. Computer vision didn't just enter basketball. It democratized it.

zostaff

231,762 次观看 • 6 个月前

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.

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 • 2 个月前

I built something similar last year with Claude, but I had to understand homography, trajectory estimation, and temporal event detection, plus stitch together detection and pose-estimation models and train a few myself. Even then, ball detection was shaky. I also tried Gemini VLMs on match clips, but the cost was too high for my use case. Now Astra appears to handle much of this zero-shot, no expertise required from developers. Wild how quickly things are moving. 😅 See an early version of my tennis ai app below.

I built something similar last year with Claude, but I had to understand homography, trajectory estimation, and temporal event detection, plus stitch together detection and pose-estimation models and train a few myself. Even then, ball detection was shaky. I also tried Gemini VLMs on match clips, but the cost was too high for my use case. Now Astra appears to handle much of this zero-shot, no expertise required from developers. Wild how quickly things are moving. 😅 See an early version of my tennis ai app below.

tian/天

12,247 次观看 • 1 个月前

🏀 I built this PoC using a single 2D broadcast video: applied pitch detection, player positioning, and movement tracking to reconstruct the play in 3D and map it onto a real basketball court in XR.

🏀 I built this PoC using a single 2D broadcast video: applied pitch detection, player positioning, and movement tracking to reconstruct the play in 3D and map it onto a real basketball court in XR.

Stijn Spanhove

36,957 次观看 • 11 个月前

We’ve upgraded Stable Video Diffusion 4D to Stable Video 4D 2.0 (SV4D 2.0), improving the quality of 4D outputs generated from a single object-centric video. While 3D provides a static view of an object’s shape and size; 4D extends this by including time, showing how the object moves. This multi-view video diffusion model generates a 4D output in three steps: 1️⃣ Starts with an input video of a moving person or object 2️⃣ Generates novel views of the subject from unseen angles 3️⃣ Constructs a single dynamic 4D video output with spatial and temporal consistency You can learn more here: (1/4)

We’ve upgraded Stable Video Diffusion 4D to Stable Video 4D 2.0 (SV4D 2.0), improving the quality of 4D outputs generated from a single object-centric video. While 3D provides a static view of an object’s shape and size; 4D extends this by including time, showing how the object moves. This multi-view video diffusion model generates a 4D output in three steps: 1️⃣ Starts with an input video of a moving person or object 2️⃣ Generates novel views of the subject from unseen angles 3️⃣ Constructs a single dynamic 4D video output with spatial and temporal consistency You can learn more here: (1/4)

Stability AI

36,110 次观看 • 1 年前

Added occlusion handling to my ball tracking pipeline ⚽️ Now a ball can't just disappear mid-play - Predicts position using velocity when ball is hidden - Merges broken tracks Tested on a longer video - the heatmap effect is very satisfying ♨️

Added occlusion handling to my ball tracking pipeline ⚽️ Now a ball can't just disappear mid-play - Predicts position using velocity when ball is hidden - Merges broken tracks Tested on a longer video - the heatmap effect is very satisfying ♨️

mirrash.eth 🦇🔊

25,029 次观看 • 9 个月前

here's a follow up ground based drone tracker :> this little setup runs with 100Hz+ tracking updates, ~20ms latency, low light, no motion blur. (4d object tracking + 6dof pose/velocity)

here's a follow up ground based drone tracker :> this little setup runs with 100Hz+ tracking updates, ~20ms latency, low light, no motion blur. (4d object tracking + 6dof pose/velocity)

jaisel

116,960 次观看 • 6 个月前

AI TENNIS ANALYSIS. A FULL COMPUTER VISION SYSTEM. BUILT ON YOLO, PYTORCH, AND KEYPOINT EXTRACTION. Take any tennis match broadcast, any camera angle, any resolution. Feed it into the pipeline. YOLO detects both players and the tennis ball frame by frame. No manual labeling, no pre-annotated dataset. A fine-tuned YOLOv5 model trained on a Roboflow tennis ball dataset handles the ball - the hardest object to track in any sport. Tiny, fast, constantly occluded. The model finds it anyway. Trackers maintain identity across frames so Player 1 stays Player 1 from the first serve to match point. But detection is just the start. A ResNet50 CNN trained in PyTorch predicts court keypoints from every frame - the corners, service lines, baselines, net posts. Fourteen points that define the entire playing surface geometry. From those keypoints the system builds a homography matrix and warps the broadcast perspective into a top-down mini court with real coordinates. Now every player has a position in real space, not pixel space. Every frame becomes a measurement. Every rally becomes a dataset. Player movement speed - calculated from position deltas between frames, converted to meters per second through the homography. Ball shot speed - measured from the ball trajectory across consecutive detections. Number of shots per rally - counted automatically through ball direction changes. All of this rendered live on the video as an overlay. A mini court in the corner showing both players as dots moving in real time. Stats updating after every point. OpenCV handles the rendering. Pandas handles the math. PyTorch handles the intelligence. YOLO handles the eyes. No Hawkeye subscription, no court-embedded sensors, no tracking chips in the ball. A Python script, a trained model, and a GPU. The full code is on GitHub. The tutorial walks through every module - from ball detector training to court keypoint extraction to the final statistical overlay. Professional teams used to need broadcast deals and proprietary hardware for this kind of analysis. Now you build it in an afternoon with open-source tools. Trading here: Computer vision didn't just enter tennis. It made the expensive stuff free.

AI TENNIS ANALYSIS. A FULL COMPUTER VISION SYSTEM. BUILT ON YOLO, PYTORCH, AND KEYPOINT EXTRACTION. Take any tennis match broadcast, any camera angle, any resolution. Feed it into the pipeline. YOLO detects both players and the tennis ball frame by frame. No manual labeling, no pre-annotated dataset. A fine-tuned YOLOv5 model trained on a Roboflow tennis ball dataset handles the ball - the hardest object to track in any sport. Tiny, fast, constantly occluded. The model finds it anyway. Trackers maintain identity across frames so Player 1 stays Player 1 from the first serve to match point. But detection is just the start. A ResNet50 CNN trained in PyTorch predicts court keypoints from every frame - the corners, service lines, baselines, net posts. Fourteen points that define the entire playing surface geometry. From those keypoints the system builds a homography matrix and warps the broadcast perspective into a top-down mini court with real coordinates. Now every player has a position in real space, not pixel space. Every frame becomes a measurement. Every rally becomes a dataset. Player movement speed - calculated from position deltas between frames, converted to meters per second through the homography. Ball shot speed - measured from the ball trajectory across consecutive detections. Number of shots per rally - counted automatically through ball direction changes. All of this rendered live on the video as an overlay. A mini court in the corner showing both players as dots moving in real time. Stats updating after every point. OpenCV handles the rendering. Pandas handles the math. PyTorch handles the intelligence. YOLO handles the eyes. No Hawkeye subscription, no court-embedded sensors, no tracking chips in the ball. A Python script, a trained model, and a GPU. The full code is on GitHub. The tutorial walks through every module - from ball detector training to court keypoint extraction to the final statistical overlay. Professional teams used to need broadcast deals and proprietary hardware for this kind of analysis. Now you build it in an afternoon with open-source tools. Trading here: Computer vision didn't just enter tennis. It made the expensive stuff free.

zostaff

121,478 次观看 • 6 个月前

Thats awesome! A developer used Al-powered 4D Gaussian Splatting to convert flat video footage into dynamic 3D spatial scenes. The system reconstructs different camera angles and depth information from ordinary footage, making it possible to navigate scenes in three dimensions. We're moving from recording video to digitally recreating reality itself.

Thats awesome! A developer used Al-powered 4D Gaussian Splatting to convert flat video footage into dynamic 3D spatial scenes. The system reconstructs different camera angles and depth information from ordinary footage, making it possible to navigate scenes in three dimensions. We're moving from recording video to digitally recreating reality itself.

SciTech Era

73,757 次观看 • 5 个月前

Someone built an interactive GPT visualizer that shows every single step of how a model generates output. Not just the final answer. The actual pipeline.

Someone built an interactive GPT visualizer that shows every single step of how a model generates output. Not just the final answer. The actual pipeline.

Aman

107,278 次观看 • 8 个月前

DimensionX Create Any 3D and 4D Scenes from a Single Image with Controllable Video Diffusion

DimensionX Create Any 3D and 4D Scenes from a Single Image with Controllable Video Diffusion

AK

58,273 次观看 • 2 年前

"near real-time method for 6-DoF tracking of an unknown object from a monocular RGBD video sequence [...] a Neural Object Field that is learned concurrently with a pose graph optimization process in order to robustly accumulate information into a consistent 3D representation"

"near real-time method for 6-DoF tracking of an unknown object from a monocular RGBD video sequence [...] a Neural Object Field that is learned concurrently with a pose graph optimization process in order to robustly accumulate information into a consistent 3D representation"

Fabien Benetou

222,962 次观看 • 3 年前

4DGS from a single camera. 4DAnyone Create Anyone in 4D from a Casual Monocular Video (SIGGRAPH Asia) Project: Code (Apache 2.0): Model:

4DGS from a single camera. 4DAnyone Create Anyone in 4D from a Casual Monocular Video (SIGGRAPH Asia) Project: Code (Apache 2.0): Model:

Radiance Fields

25,504 次观看 • 2 个月前

🚨 BRAVE & STRONG: Ann Widdecombe: “I am prepared to depart from a path on which I was convinced my life was set, because democracy is more important than any single individual.”

🚨 BRAVE & STRONG: Ann Widdecombe: “I am prepared to depart from a path on which I was convinced my life was set, because democracy is more important than any single individual.”

REFORM DAILY WATCH 🏴󠁧󠁢󠁥󠁮󠁧󠁿🇬🇧

22,689 次观看 • 5 个月前