Research map / Video / Video segmentation and tracking
AssembleNet: Searching for Multi-Stream Neural Connectivity in Video Architectures
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use connection-learning-guided evolution instead of random architecture search or random connection mutations (ablation) | AssembleNet architecture search · Moments in Time | top-1 + top-5 accuracy fitness | Helped |
| Use random connectivity instead of the evolved connectivity (ablation) | Multi-stream video CNN · Moments in Time and Charades | classification performance | Hurt |
| Use hand-designed Flow→RGB connectivity instead of evolved AssembleNet connectivity (ablation) | Multi-stream (2+1)D CNN · Moments in Time | accuracy | Hurt |
| Use hand-designed Flow→RGB connectivity instead of evolved AssembleNet connectivity (ablation) | Multi-stream (2+1)D CNN · Charades | mAP: 49.5 -> 53.0 | Hurt |
| Evolve multi-stream connectivity rather than use a two-stream (2+1)D ResNet-50 | Charades_v1_classify, Kinetics pre-training · Charades_v1_classify | mAP: 50.4 -> 56.6 | Helped |
| Evolve multi-stream connectivity rather than use a two-stream (2+1)D ResNet-50 | Charades_v1_classify, no pre-training · Charades_v1_classify | mAP: 39.9 -> 47.0 | Helped |
Closest papers by meaning
- Convolutional Two-Stream Network Fusion for Video Action Recognition (CVPR 2016 · similarity 0.87)
- Spatiotemporal Residual Networks for Video Action Recognition (NeurIPS 2016 · similarity 0.87)
- Cream of the Crop: Distilling Prioritized Paths For One-Shot Neural Architecture Search (NeurIPS 2020 · similarity 0.87)
- Spatiotemporal Pyramid Network for Video Action Recognition (CVPR 2017 · similarity 0.87)
- More Is Less: Learning Efficient Video Representations by Big-Little Network and Depthwise Temporal Aggregation (NeurIPS 2019 · similarity 0.87)
- Video Modeling With Correlation Networks (CVPR 2020 · 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.