Research map / Detection and segmentation / Open-set and few-shot detection
SOD-MTGAN: Small Object Detection via Multi-Task Generative Adversarial Network
Yancheng Bai, Yongqiang Zhang, Mingli Ding, Bernard Ghanem
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 |
|---|---|---|---|
| add regression branch to the discriminator (ablation) | SOD-MTGAN · COCO minival | APS: 19.0 -> 19.9 | Improved |
| add regression branch to the discriminator (ablation) | SOD-MTGAN · COCO minival | AP: 36.7 -> 37.2 | Improved |
| add adversarial loss to the generator (ablation) | SOD-MTGAN · COCO minival | APS: 19.4 -> 19.9 | Improved |
| add classification loss to the generator (ablation) | SOD-MTGAN · COCO minival | APS: 19.2 -> 19.9 | Improved |
| add regression loss to the generator (ablation) | SOD-MTGAN · COCO minival | APS: 19.1 -> 19.9 | Improved |
| add adversarial loss to the generator (ablation) | SOD-MTGAN · COCO minival | AP: 37.0 -> 37.2 | Improved |
| add MTGAN super-resolution stage on ResNet-50-C4 detector (ablation) | Faster-RCNN · COCO minival | APS: 18.4 -> 19.9 | Improved |
| add MTGAN super-resolution stage on ResNet-101-FPN detector (ablation) | Mask-RCNN / FPN · COCO minival | APS: 23.5 -> 25.1 | Improved |
| add MTGAN stage to baseline detector, overall AP (ResNet-50) (ablation) | Faster-RCNN · COCO minival | AP: 36.5 -> 37.2 | Improved |
| add MTGAN stage to baseline detector, overall AP (ResNet-101) (ablation) | Mask-RCNN · COCO minival | AP: 40.9 -> 41.5 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Faster-RCNN — Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks (TPAMI 2016) · better on 6 of 6 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Perceptual Generative Adversarial Networks for Small Object Detection (CVPR 2017 · similarity 0.93)
- Finding Tiny Faces in the Wild With Generative Adversarial Network (CVPR 2018 · similarity 0.89)
- Better to Follow, Follow to Be Better: Towards Precise Supervision of Feature Super-Resolution for Small Object Detection (ICCV 2019 · similarity 0.87)
- Multi-task Generative Adversarial Network for Detecting Small Objects in the Wild (IJCV 2020 · similarity 0.87)
- A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection (ECCV 2016 · similarity 0.87)
- Generative Adversarial Learning Towards Fast Weakly Supervised Detection (CVPR 2018 · similarity 0.87)
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.
