EgoM2P: Egocentric Multimodal Multitask Pretraining
cs.CV
Submitted: 2025-06-09
Updated: 2026-11-04
Project page: https://egom2p.github.io/Abstract
Terminology
Sources
- Cosmos World Foundation Model Platform for Physical AI
- Introducing HOT3D: An Egocentric Dataset for 3D Hand and Object Tracking
- An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
- VIOLET : End-to-End Video-Language Transformers with Masked Visual-token Modeling
- World Models
- GPT-4o System Card
- VideoChat: Chat-Centric Video Understanding
- Aria Everyday Activities Dataset
- Nymeria: A Massive Collection of Multimodal Egocentric Daily Motion in the Wild
- Transformers are Sample-Efficient World Models
- Movie Gen: A Cast of Media Foundation Models
- The Replica Dataset: A Digital Replica of Indoor Spaces
- Gemini: A Family of Highly Capable Multimodal Models
- Diffusion Models Are Real-Time Game Engines
- EgoVid-5M: A Large-Scale Video-Action Dataset for Egocentric Video Generation
- InternVideo: General Video Foundation Models via Generative and Discriminative Learning
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs