Research map / Detection and segmentation / Open-set and few-shot detection
DropLoss for Long-Tail Instance Segmentation
Ting-I Hsieh, Esther Robb, Hwann-Tzong Chen, Jia-Bin Huang
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 |
|---|---|---|---|
| Replace BCE with DropLoss | Mask R-CNN, R-50-FPN · LVIS v0.5 val | AP (%): 21.5 -> 25.5 | Improved |
| Replace EQL with DropLoss | Mask R-CNN, R-50-FPN · LVIS v0.5 val | AP (%): 23.8 -> 25.5 | Improved |
| Replace EQL with DropLoss | Mask R-CNN, R-50-FPN · LVIS v0.5 val | AP_r: 8.5 -> 13.2 | Improved |
| Replace EQL with DropLoss | Mask R-CNN, R-50-FPN · LVIS v0.5 val | AP_f: 28.3 -> 27.3 | Hurt |
| Replace EQL with DropLoss | Mask R-CNN, R-101-FPN · LVIS v0.5 val | AP (%): 26.2 -> 26.9 | Improved |
| Replace EQL with DropLoss | Cascade R-CNN, R-50-FPN · LVIS v0.5 val | AP (%): 24.2 -> 25 | Improved |
| Replace EQL with DropLoss | Cascade R-CNN, R-101-FPN · LVIS v0.5 val | AP (%): 25.4 -> 26.4 | Improved |
| Add Repeat Factor Sampling | DropLoss with Mask R-CNN, ResNet-50 · LVIS v0.5 val | AP (%): 25.5 -> 26.4 | Improved |
| Add Repeat Factor Sampling | DropLoss with Mask R-CNN, ResNet-50 · LVIS v1.0 val | AP (%): 19.8 -> 22.3 | Improved |
| Use BEQL with logarithm base b=4 instead of DropLoss (ablation) | Mask R-CNN, ResNet-50 · LVIS v0.5 val | AP (%): 25.5 -> 25.2 | Hurt |
| Use BEQL with logarithm base b=5 instead of DropLoss (ablation) | Mask R-CNN, ResNet-50 · LVIS v0.5 val | AP (%): 25.5 -> 25.1 | Hurt |
Closest papers by meaning
- The Devil is in Classification: A Simple Framework for Long-tail Instance Segmentation (ECCV 2020 · similarity 0.89)
- Seesaw Loss for Long-Tailed Instance Segmentation (CVPR 2021 · similarity 0.88)
- Learning with Free Object Segments for Long-Tailed Instance Segmentation (ECCV 2022 · similarity 0.88)
- Relieving Long-Tailed Instance Segmentation via Pairwise Class Balance (CVPR 2022 · similarity 0.88)
- Adaptive Class Suppression Loss for Long-Tail Object Detection (CVPR 2021 · similarity 0.87)
- Pursuing Better Decision Boundaries for Long-Tailed Object Detection via Category Information Amount (ICLR 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.
