Research map / Video / Video segmentation and tracking
Unsupervised Learning for Physical Interaction through Video Prediction
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Predict pixel motion by transforming pixels from previous frames instead of reconstructing frames from scratch (ablation) | Video prediction models · Robotic pushing dataset | PSNR and SSIM | Helped |
| Predict pixel motion instead of reconstructing raw pixel values or predicting frame differences (ablation) | Stacked convolutional LSTM · Robotic pushing dataset, including novel objects | video prediction accuracy | Helped |
| Use scheduled sampling during training (ablation) | Proposed recurrent models · Robotic pushing dataset | validation error | Helped |
| Use scheduled sampling during training (ablation) | Ablation and baseline models that do not model pixel motion · Robotic pushing dataset | validation error | No clear effect |
| Train for longer prediction horizons (ablation) | Proposed model · Robotic pushing dataset | video prediction accuracy | Helped |
| Condition predictions on the robot's actions (ablation) | Proposed model · Robotic pushing dataset | action-conditioned prediction quality | Helped |
| Use pixel-motion prediction instead of prior video prediction methods | DNA, CDNA, and STP · Human3.6M held-out subject | video prediction accuracy | Helped |
Closest papers by meaning
- PlaySlot: Learning Inverse Latent Dynamics for Controllable Object-Centric Video Prediction and Planning (ICML 2025 · similarity 0.91)
- ViPRA: Video Prediction for Robot Actions (ICLR 2026 · similarity 0.90)
- Learning an Actionable Discrete Diffusion Policy via Large-Scale Actionless Video Pre-Training (NeurIPS 2024 · similarity 0.88)
- Video Prediction Policy: A Generalist Robot Policy with Predictive Visual Representations (ICML 2025 · similarity 0.88)
- Object-centric 3D Motion Field for Robot Learning from Human Videos (NeurIPS 2025 · similarity 0.88)
- Robots Pre-train Robots: Manipulation-Centric Robotic Representation from Large-Scale Robot Datasets (ICLR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.