Research map / Video / Video segmentation and tracking
Learning Disentangled Representations of Videos with Missing Data
Armand Comas, Chi Zhang, Zlatan Feric, Octavia Camps, Rose Yu
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 |
|---|---|---|---|
| add missingness variable and imputation (ablation) | DIVE · Moving MNIST S2 | BCE (reconstruction): 236.35 -> 165.42 | Improved |
| add missingness variable and imputation (ablation) | DIVE · Moving MNIST S2 | MSE (reconstruction): 49.07 -> 27.03 | Improved |
| add missingness variable and imputation (ablation) | DIVE · Moving MNIST S2 | PSNR (prediction): 17.66 -> 18.56 | Improved |
| add missingness variable and imputation (ablation) | DIVE · Moving MNIST S2 | SSIM (prediction): 0.82 -> 0.83 | Improved |
| DIVE vs DDPAE | DIVE · Moving MNIST S1 | BCE (prediction): 417.00 -> 459.10 | Hurt |
| DIVE vs DDPAE | DIVE · Moving MNIST S1 | PSNR (reconstruction): 17.56 -> 20.64 | Improved |
| DIVE vs DDPAE | DIVE · Moving MNIST S2 | BCE (prediction): 409.26 -> 321.29 | Improved |
| DIVE vs DDPAE | DIVE · Moving MNIST S3 | BCE (prediction): 403.48 -> 374.59 | Improved |
| DIVE vs DDPAE | DIVE · Moving MNIST S3 | SSIM (prediction): 0.78 -> 0.78 | No clear effect |
| DIVE vs DDPAE | DIVE · MOTS pedestrians | BCE: 2495.08 -> 1355.89 | Improved |
| DIVE vs DDPAE | DIVE · MOTS pedestrians | SSIM: 0.90 -> 0.96 | Improved |
| soft sigmoid missingness labels instead of hard heaviside labels (ablation) | DIVE · Moving MNIST | performance | No clear effect |
Closest papers by meaning
- Video Autoencoder: Self-Supervised Disentanglement of Static 3D Structure and Motion (ICCV 2021 · similarity 0.89)
- Learning to Decompose and Disentangle Representations for Video Prediction (NeurIPS 2018 · similarity 0.89)
- Factorized Diffusion Autoencoder for Unsupervised Disentangled Representation Learning (AAAI 2024 · similarity 0.86)
- Generative Omnimatte: Learning to Decompose Video into Layers (CVPR 2025 · similarity 0.86)
- Disentanglement Beyond Static vs. Dynamic: A Benchmark and Evaluation Framework for Multi-Factor Sequential Representations (NeurIPS 2025 · similarity 0.86)
- EVEREST: Efficient Masked Video Autoencoder by Removing Redundant Spatiotemporal Tokens (ICML 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.
