MoRAE: Flow-Friendly Self-Supervised Latents for Text-to-Motion Generation
cs.CV, cs.AI
Submitted: 2026-07-31
Updated: 2026-11-15
Code: https://github.com/zhuyifeiabcd1/MoRAE
Terminology
Sources
- Emerging Properties in Self-Supervised Vision Transformers
- Generative Human Motion Stylization in Latent Space
- Masked Autoencoders Are Scalable Vision Learners
- Rethinking Diffusion for Text-Driven Human Motion Generation: Redundant Representations, Evaluation, and Masked Autoregression
- DINOv2: Learning Robust Visual Features without Supervision
- Improved Baselines with Representation Autoencoders
- MotionDiffuse: Text-Driven Human Motion Generation with Diffusion Model
- Diffusion Transformers with Representation Autoencoders
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