Research map / Video / Video segmentation and tracking
VideoTitans: Scalable Video Prediction with Integrated Short- and Long-term Memory
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 VideoTitans, a video prediction model that integrates short- and long-term memory to forecast high-resolution, long-horizon video with scalable complexity.
Problem
Video forecasting for high-resolution, long-horizon prediction, where transformers' global attention has quadratic cost and convolutional/recurrent methods are limited.
Approach
Integrates short-term and long-term memory modules into a video prediction architecture to avoid quadratic global attention while retaining long-range context (details beyond the abstract not available).
Closest papers by meaning
- Learning to Generate Long-term Future via Hierarchical Prediction (ICML 2017 · similarity 0.87)
- VEDIT: Latent Prediction Architecture For Procedural Video Representation Learning (ICLR 2025 · similarity 0.87)
- MESSENGER: Memory-Enhanced Sequential Scene Flow Estimation via Autoregressive Next-Frame Forecasting (NeurIPS 2026 · similarity 0.86)
- A Frame is Worth One Token: Efficient Generative World Modeling with Delta Tokens (CVPR 2026 · similarity 0.86)
- StreamForest: Efficient Online Video Understanding with Persistent Event Memory (NeurIPS 2025 · similarity 0.86)
- VisionTS: Visual Masked Autoencoders Are Free-Lunch Zero-Shot Time Series Forecasters (ICML 2025 · 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.