Research map / Vision-language and multimodal / Multimodal LLMs
Placepedia: Comprehensive Place Understanding with Multi-Faceted Annotations
Huaiyi Huang, Yuqi Zhang, Qingqiu Huang, Zhengkui Guo, Ziwei Liu, Dahua Lin
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 on the larger Places-Coarse set | PlaceNet · place retrieval top-1 | accuracy: 54.31 -> 67.85 | Improved |
| train on the larger Places-Coarse set | PlaceNet · city recognition top-1 | accuracy: 18.73 -> 29.25 | Improved |
| replace average pooling with max pooling (ablation) | PlaceNet · place retrieval top-1 | accuracy: 54.33 -> 49.66 | Hurt |
| replace average pooling with spatial pyramid pooling (ablation) | PlaceNet · place retrieval top-1 | accuracy: 54.33 -> 28.18 | Hurt |
| replace average pooling with spatial pyramid pooling (ablation) | PlaceNet · place categorization top-1 | accuracy: 25.95 -> 27.21 | Improved |
| add triplet loss to softmax (ablation) | PlaceNet · place retrieval top-1 | accuracy: 54.31 -> 50.33 | Hurt |
| add triplet loss to softmax (ablation) | PlaceNet · place categorization top-1 | accuracy: 25.95 -> 21.15 | Hurt |
| replace softmax with focal loss (ablation) | PlaceNet · place retrieval top-1 | accuracy: 54.31 -> 55.03 | Improved |
| replace softmax with focal loss (ablation) | PlaceNet · city recognition top-1 | accuracy: 18.73 -> 18.67 | No clear effect |
| use ResNet50 instead of AlexNet backbone | PlaceNet baseline · place retrieval top-1 | accuracy: 33.78 -> 54.53 | Improved |
| use ResNet50 instead of VGG16 backbone | PlaceNet baseline · function categorization top-1 | accuracy: 65.97 -> 68.25 | Improved |
| use ResNet50 instead of VGG16 backbone | PlaceNet baseline · country recognition top-1 | accuracy: 24.86 -> 19.72 | Hurt |
Closest papers by meaning
- Places: A 10 Million Image Database for Scene Recognition (TPAMI 2017 · similarity 0.90)
- 24/7 Place Recognition by View Synthesis (TPAMI 2017 · similarity 0.87)
- SAGE: Spatial-visual Adaptive Graph Exploration for Efficient Visual Place Recognition (ICLR 2026 · similarity 0.87)
- Towards Implicit Aggregation: Robust Image Representation for Place Recognition in the Transformer Era (NeurIPS 2025 · similarity 0.85)
- Focus on Local: Finding Reliable Discriminative Regions for Visual Place Recognition (AAAI 2025 · similarity 0.85)
- BoQ: A Place is Worth a Bag of Learnable Queries (CVPR 2024 · similarity 0.85)
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.
