Research map / 3D vision / 3D generation
Good View Hunting: Learning Photo Composition From Dense View Pairs
Zijun Wei, Jianming Zhang, Xiaohui Shen, Zhe Lin, Radomír Mech, Minh Hoai, Dimitris Samaras
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| Train the view evaluation model on CPC comparative pairs with ranking loss | VEN · XPView test | swap error: 0.28 -> 0.22 | Improved |
| Use ranking loss rather than classification loss on CPC annotations (ablation) | VEN · XPView test | swap error: 0.24 -> 0.22 | Improved |
| Use MPSE instead of MSE for teacher-to-student transfer (ablation) | VPN · 895 anchor boxes | VPN-VEN swap error: 0.1601 -> 0.1554 | Improved |
| Add 40K AADB images to the CPC transfer set (ablation) | VPN · 895 anchor boxes | VPN-VEN swap error: 0.1554 -> 0.1312 | Improved |
| Use teacher-guided knowledge transfer rather than direct anchor-box classification | VPN · FCDB | IoU: 0.6429 -> 0.7109 | Improved |
| Use teacher-guided knowledge transfer rather than direct anchor-box classification | VPN · FLMS | IoU: 0.7875 -> 0.8352 | Improved |
| Use teacher-guided knowledge transfer rather than direct anchor-box classification | VPN · Fast-AT thumbnail test | IoU | Improved |
| Use CPC comparative annotations rather than AVA image-level aesthetics labels | VEN · XPView test | swap error: 0.37 -> 0.22 | Improved |
Closest papers by meaning
- Can Machines Understand Composition? Dataset and Benchmark for Photographic Image Composition Embedding and Understanding (CVPR 2025 · similarity 0.86)
- SELECT: A Large-Scale Benchmark of Data Curation Strategies for Image Classification (NeurIPS 2024 · similarity 0.84)
- Learning Visual Composition through Improved Semantic Guidance (CVPR 2025 · similarity 0.84)
- Compositional Scene Understanding through Inverse Generative Modeling (ICML 2025 · similarity 0.84)
- Instance-Level Composed Image Retrieval (NeurIPS 2025 · similarity 0.84)
- Photography Perspective Composition: Towards Aesthetic Perspective Recommendation (NeurIPS 2025 · similarity 0.84)
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.
