Shaowei Liu

📢MoRight: Motion Control Done Right "What if your video model actually understood cause and effect?" Existing motion-controlled video models entangle camera and object motion, and treat everything as kinematic displacement. MoRight changes both. 🔥 Motion Causality — MoRight decomposes motion into actions & consequences. Give an action → MoRight predicts consequences (aka motion simulation) . Give a desired outcome → MoRight recovers the driving action (aka motion planning). Not merely displacing pixels. 🎬 Disentangled Control — MoRight separates camera and object motion, allowing users to independently control each of them. No entanglement. Project Page: research.nvidia.com/labs/sil/proje… Paper: arxiv.org/abs/2604.07348

What is missing to bring real-time motion research into AAA games and real-world robotics? We present MotionBricks, a step toward bridging this gap with two key components: - a single generative latent motion backbone covering 350,000+ motion skills, running at 15,000 FPS with 2 ms latency and substantially improved quality and reliability. - a unified smart primitive interface for locomotion, object / scene interaction, with fine-grained control over generated behaviors. Webpage: nvlabs.github.io/motionbricks/ Code: github.com/NVlabs/GR00T-W… Paper: arxiv.org/abs/2604.24833 (ACM TOG / SIGGRAPH 2026)

We scaled up Lyra to generate explorable 3D worlds! 🚀 Introducing Lyra 2.0 — turning a single image into a 3D world you can walk through, look back, and even drop a robot into 🤖 Code and Model available today! 🌐 Website: research.nvidia.com/labs/sil/proje… (1/N)

@Yahskapar Thanks, Akshay.

@stevenpg8 Very interesting work - congrats and great job!

📢MoRight: Motion Control Done Right "What if your video model actually understood cause and effect?" Existing motion-controlled video models entangle camera and object motion, and treat everything as kinematic displacement. MoRight changes both. 🔥 Motion Causality — MoRight decomposes motion into actions & consequences. Give an action → MoRight predicts consequences (aka motion simulation) . Give a desired outcome → MoRight recovers the driving action (aka motion planning). Not merely displacing pixels. 🎬 Disentangled Control — MoRight separates camera and object motion, allowing users to independently control each of them. No entanglement. Project Page: research.nvidia.com/labs/sil/proje… Paper: arxiv.org/abs/2604.07348

MoRight tackles a core problem in controllable video generation: disentangling camera motion from object motion. Unlike prior methods that rely on dense future-frame tracks, MoRight uses only first-frame reprojected trajectories + camera poses, yet achieves comparable quality and better motion control.

When doing motion-conditioned video generation, we think the key is in understanding motion causality (what action will lead to what outcome) and camera-motion decomposition (camera changes shouldn't entangle with object dynamics). MoRight is the first step towards this goal. We support three functionalities: 1. "Simulation": Users provide the action (e.g., moving hands), and the video model generates the consequences (e.g., cups moving and pouring water) 2. "Planning": Users specify the outcome (e.g., balls moving), and the video model generates the action that drives the outcome (e.g., moving legs) 3. "Disentanglement": Users can independently control the camera change and object dynamics, all along with the motion causality mentioned above. Come and check it at research.nvidia.com/labs/sil/proje… P.S. Shaowei is in the job market; you shouldn't miss him!

The project is a joint effort with @xuanchi13 @TianchangS @HuanLing6 @_saurabhg @ShenlongWang @FidlerSanja @JunGao33210520 @nvidia and @UofIllinois, thanks everyone! Special thanks to Spatial Intelligence Lab for hosting me as a intern! Check more results at 👇research.nvidia.com/labs/sil/proje…

Our method is straightforward! Dual-stream architecture: Dual-stream architecture: one stream for object motion on the canonical frame, one for camera motion. Object motion transferred across views via cross-view self-attention. Data: We curate a systematic data pipeline to decompose the motion sequences in our training video into active and passive motion, allowing the model to learn to capture the motion causality.

Visit  stevenlsw.github.io/visualsync and try the cool demo at github.com/stevenlsw/visu… Don’t miss our poster session tmr! Poster: Thur Dec. 3, 11-2pm #4607 Joint work with @david_yao14255, @_saurabhg, @ShenlongWang and @siebelschool Come and unlock more fun demos at our poster tmr👇

Our method is generic, robust and scalable:  we test on ego-exo settings, sports game recordings, multi-person activities, animals. The synched outputs  benefit dynamic reconstruction, novel view synthesis, and multi-view data engines. Check demos and applications on our website

Come and visit #NeurIPS2025 work tmr at #4607: VisualSync: Multi-Camera Synchronization via Cross-View Object Motion Website: stevenlsw.github.io/visualsync/ Paper: arxiv.org/pdf/2512.02017 Code: github.com/stevenlsw/visu… Poster:  11-2pm pst #4607 Multi-view (>=30) videos sync in-the-wild

Then our method is straightforward: we need cross-view temporal correspondences and camera pose to compute epipolar lines. We estimate pairwise video offsets by minimizing epipolar violations over matched correspondences. Later we perform global optimization to align all videos.

In a dynamic scene with moving camera, the epipolar geometry still holds: In synced videos (left), the green correspondence from one view always align with the red epipolar lines in the other view. In unsynced videos (right), deviations from the red lines indicate the sync error.

@LeaMue27 @camenduru @chuan_guo92603 @_JianWang_ Thanks so much Lea, your amazing pioneer work inspires us a lot

@stevenpg8 @camenduru @chuan_guo92603 @_JianWang_ Cool work! I love how these interactions come to life in 4D.

Glad introduce our #ICCV2025 work Ponimator: Unfolding Interactive Pose for Versatile Human-human Interaction Animation Website:  stevenlsw.github.io/ponimator/ Paper: arxiv.org/abs/2510.14976 Code: stevenlsw.github.io/ponimator/ Poster:  Wed 10.22, 10:30am #1109 @chuan_guo92603 @_JianWang_

Visit stevenlsw.github.io/ponimator and try the cool demo at github.com/stevenlsw/poni…   Don’t miss our poster session on Wed Oct. 22 10:30 #1109 Joint work @chuan_guo92603, Bing Zhou, @_JianWang_,  @Snapchat and @IllinoisCDS Warm hug from the most famous statue at Waikiki beach👇

Our framework enables diverse applications, including creating vivid interactive videos between people from 2 single image like below👇