Research map / Video / Video generation
Probabilistic Video Generation using Holistic Attribute Control
Jiawei He, Andreas Lehrmann, Joseph Marino, Greg Mori, Leonid Sigal
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 an LSTM temporal model to a static VAE (ablation) | VAE · Chair CAD | Intra-E: 40.33 -> 17.64 | Improved |
| add an LSTM temporal model to a static VAE (ablation) | VAE · Chair CAD | I-Score: 1.28 -> 1.83 | Improved |
| add structured latent space with holistic attribute control (ablation) | VAE+LSTM · Chair CAD | Intra-E: 17.64 -> 7.79 | Improved |
| add structured latent space with holistic attribute control (ablation) | VAE+LSTM · Chair CAD | I-Score: 1.83 -> 3.63 | Improved |
| add conditional sampling at each time step (ablation) | VAE+LSTM · Chair CAD | Intra-E: 17.64 -> 14.81 | Improved |
| add conditional sampling at each time step (ablation) | VAE+LSTM · Chair CAD | I-Score: 1.83 -> 2.56 | Improved |
| add conditional sampling on top of structured latent space (ablation) | VideoVAE · Chair CAD | Intra-E: 7.79 -> 5.50 | Improved |
| add structured latent space on top of conditional sampling (ablation) | VideoVAE · Chair CAD | I-Score: 2.56 -> 3.94 | Improved |
| use VideoVAE instead of MoCoGAN | VideoVAE · Weizmann Human Action | I-Score: 13.87 -> 69.55 | Improved |
| use VideoVAE instead of Deep Rotator | VideoVAE · Chair CAD | Intra-E: 14.68 -> 5.50 | Improved |
| use VideoVAE instead of VGAN | VideoVAE · YFCC MIT Flickr | I-Score: 1.58 -> 1.62 | Improved |
| condition on first frame (prediction) instead of de-novo synthesis | VideoVAE · YFCC MIT Flickr | Intra-E: 44.03 -> 38.20 | Improved |
Closest papers by meaning
- Generating Long Videos of Dynamic Scenes (NeurIPS 2022 · similarity 0.89)
- Stochastic Video Generation with a Learned Prior (ICML 2018 · similarity 0.88)
- High-Quality Joint Image and Video Tokenization with Causal VAE (ICLR 2025 · similarity 0.88)
- CAGE: Unsupervised Visual Composition and Animation for Controllable Video Generation (AAAI 2025 · similarity 0.86)
- CV-VAE: A Compatible Video VAE for Latent Generative Video Models (NeurIPS 2024 · similarity 0.86)
- WorldWeaver: Generating Long-Horizon Video Worlds via Rich Perception (NeurIPS 2025 · similarity 0.86)
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.
