Rethinking Asymmetry in Self-Supervised 3D Geometry Learning
cs.CV
Submitted: 2026-10-30
Updated: 2026-10-30
Terminology
Sources
- Self-Supervised Flow Matching for Scalable Multi-Modal Synthesis
- WildRayZer: Self-supervised Large View Synthesis in Dynamic Environments
- Self-Improving 4D Perception via Self-Distillation
- From None to All: Self-Supervised 3D Reconstruction via Novel View Synthesis
- No Other Representation Component Is Needed: Diffusion Transformers Can Provide Representation Guidance by Themselves
- LVSM: A Large View Synthesis Model with Minimal 3D Inductive Bias
- Scaling Laws for Neural Language Models
- Feed-Forward Bullet-Time Reconstruction of Dynamic Scenes from Monocular Videos
- Depth Anything 3: Recovering the Visual Space from Any Views
- Decoupled Weight Decay Regularization
- DINOv2: Learning Robust Visual Features without Supervision
- RayDer: Scalable Self-Supervised Novel View Synthesis from Real-World Video
- DINOv3
- The Less You Depend, The More You Learn: Synthesizing Novel Views from Sparse, Unposed Images with Minimal 3D Knowledge
- VGGT-
- 4DGT: Learning a 4D Gaussian Transformer Using Real-World Monocular Videos
- Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think
- Stereo Magnification: Learning View Synthesis using Multiplane Images
- Efficient Universal Perception Encoder
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