Research map / Video / Video segmentation and tracking
V1-Inspired Dynamic Vision System: A Bio-Plausible Video Embedding Framework with Decoupled Shape–Color Pathways and Long-Range Spatiotemporal Perception
Beier Meng
Read the paperCheck your idea against 137,217 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 V1-DVS, a bio-plausible video embedding front end inspired by V1 with decoupled shape and color pathways and long-range spatiotemporal perception, trained end-to-end in video-text contrastive learning.
Problem
Video-text contrastive learning typically uses linear patch embeddings or evenly sliced tubelets as the ViT front end, which may be suboptimal for video understanding; the paper revisits this front end using principles of primary visual cortex (V1).
Approach
Replaces the standard linear patch/tubelet embedding with a V1-inspired front end that decouples shape and color pathways and models long-range spatiotemporal interactions, trained end-to-end with video-text contrastive objectives.
Closest papers by meaning
- HIMMEL: Hierarchical Interleaved Multi-stream Motion Encoding for Long Video Understanding (NeurIPS 2026 · similarity 0.87)
- A Simple Recipe for Contrastively Pre-training Video-First Encoders Beyond 16 Frames (CVPR 2024 · similarity 0.86)
- TTVidT: Decoupling the Temporal Axis for Efficient Motion-Centric Video Pretraining (NeurIPS 2026 · similarity 0.86)
- OneVision-Encoder: Codec-Aligned Sparsity as a Foundational Principle for Multimodal Intelligence (NeurIPS 2026 · similarity 0.86)
- Perception Encoder: The best visual embeddings are not at the output of the network (NeurIPS 2025 · similarity 0.86)
- Linear Differential Vision Transformer: Learning Visual Contrasts via Pairwise Differentials (NeurIPS 2025 · similarity 0.86)
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.
