Research map / Video / Video generation
High-Quality Joint Image and Video Tokenization with Causal VAE
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes a causal VAE tokenizer that jointly compresses images and videos into a shared latent space, handling temporal redundancy in video generation.
Problem
High-dimensional video latents make video generation hard; existing methods use image autoencoders that ignore temporal redundancy in videos.
Approach
A causal VAE for joint image and video tokenization that exploits temporal redundancy while sharing one latent space across images and videos, instead of relying on per-frame pretrained image autoencoders.
Closest papers by meaning
- Improved Video VAE for Latent Video Diffusion Model (CVPR 2025 · similarity 0.90)
- When Worse is Better: Navigating the Compression Generation Trade-off In Visual Tokenization (NeurIPS 2025 · similarity 0.89)
- Improving the Diffusability of Autoencoders (ICML 2025 · similarity 0.88)
- Epsilon-VAE: Denoising as Visual Decoding (ICML 2025 · similarity 0.88)
- Learnings from Scaling Visual Tokenizers for Reconstruction and Generation (ICML 2025 · similarity 0.88)
- CV-VAE: A Compatible Video VAE for Latent Generative Video Models (NeurIPS 2024 · 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.