Research map / Video / Video segmentation and tracking
Entropy-driven Unsupervised Keypoint Representation Learning in Videos
Ali Younes, Simone Schaub-Meyer, Georgia Chalvatzaki
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| Adding MINT w/o regularization terms (information losses only) versus baseline Transporter on CLEVRER detection | MINT w/o Reg. · CLEVRER | DOP / TOP | Improved |
| Adding regularizers (overlapping, active status, distance penalty) to MINT w/o Reg. (ablation) | MINT · CLEVRER | RAK / UAK (keypoint assignment) | Improved |
| Using the Sigmoid-derivative kernel and CUDA entropy layer instead of naive histogram computation | MINT entropy layer · implementation | GPU memory: naive >50GB | Improved |
| Replacing trained keypoints with a keypoint-based imitation policy vs CNN agent on MAGICAL | MINT keypoint representation for imitation learning · MAGICAL (MoveToRegion, MoveToCorner, MakeLine) | policy success | No clear effect |
| Removing the conditional entropy (MCE) and transportation losses is discussed qualitatively; MCE focuses attention on moving objects when keypoints are scarce | MINT · SIMITATE | qualitative tracking of static objects | Improved |
Closest papers by meaning
- Unsupervised learning of object structure and dynamics from videos (NeurIPS 2019 · similarity 0.90)
- Unsupervised Learning of Object Keypoints for Perception and Control (NeurIPS 2019 · similarity 0.88)
- Unsupervised Keypoint Learning for Guiding Class-Conditional Video Prediction (NeurIPS 2019 · similarity 0.86)
- A Large-Scale Study on Unsupervised Spatiotemporal Representation Learning (CVPR 2021 · similarity 0.86)
- SeCo: Exploring Sequence Supervision for Unsupervised Representation Learning (AAAI 2021 · similarity 0.86)
- Moving Off-the-Grid: Scene-Grounded Video Representations (NeurIPS 2024 · similarity 0.85)
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.
