Papers › InstaBoost: Boosting Instance Segmentation via Probability Map Guided Copy-Pasting
InstaBoost: Boosting Instance Segmentation via Probability Map Guided Copy-Pasting
Hao-Shu Fang, Jianhua Sun, Runzhong Wang, Minghao Gou, Yong-Lu Li, Cewu Lu
Instance segmentation requires a large number of training samples to achieve satisfactory performance and benefits from proper data augmentation. To enlarge the training set and increase the diversity, previous methods have investigated using data annotation from other domain (e.g. bbox, point) in a weakly supervised mechanism. In this paper, we present a simple, efficient and effective method to augment the training set using the existing instance mask annotations. Exploiting the pixel redundancy of the background, we are able to improve the performance of Mask R-CNN for 1.7 mAP on COCO dataset and 3.3 mAP on Pascal VOC dataset by simply introducing random jittering to objects. Furthermore, we propose a location probability map based approach to explore the feasible locations that objects can be placed based on local appearance similarity. With the guidance of such map, we boost the performance of R101-Mask R-CNN on instance segmentation from 35.7 mAP to 37.9 mAP without modifying the backbone or network structure. Our method is simple to implement and does not increase the computational complexity. It can be integrated into the training pipeline of any instance segmentation model without affecting the training and inference efficiency. Our code and models have been released at https://github.com/GothicAi/InstaBoost
In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Instance Segmentation | COCO test-dev | Cascade R-CNN (ResNet-101-FPN, map-guided) | AP50 | 61.4% | #80 of 112 | Archive leaderboard | report |
| Instance Segmentation | COCO test-dev | Cascade R-CNN (ResNet-101-FPN, map-guided) | AP75 | 42.9% | #80 of 112 | Archive leaderboard | report |
| Instance Segmentation | COCO test-dev | Cascade R-CNN (ResNet-101-FPN, map-guided) | APL | 52.1% | #80 of 112 | Archive leaderboard | report |
| Instance Segmentation | COCO test-dev | Cascade R-CNN (ResNet-101-FPN, map-guided) | APM | 42.5% | #80 of 112 | Archive leaderboard | report |
| Instance Segmentation | COCO test-dev | Cascade R-CNN (ResNet-101-FPN, map-guided) | APS | 21.2% | #80 of 112 | Archive leaderboard | report |
| Instance Segmentation | COCO test-dev | Cascade R-CNN (ResNet-101-FPN, map-guided) | mask AP | 39.5% | #80 of 112 | Archive leaderboard | report |
| Object Detection | COCO test-dev | Cascade R-CNN-FPN (ResNet-101, map-guided) | AP50 | 64.2 | #134 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | Cascade R-CNN-FPN (ResNet-101, map-guided) | AP75 | 50 | #134 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | Cascade R-CNN-FPN (ResNet-101, map-guided) | APL | 58.6 | #134 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | Cascade R-CNN-FPN (ResNet-101, map-guided) | APM | 49 | #134 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | Cascade R-CNN-FPN (ResNet-101, map-guided) | APS | 26.3 | #134 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | Cascade R-CNN-FPN (ResNet-101, map-guided) | box mAP | 45.9 | #134 of 225 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Methods
Introduced by this paper: InstaBoost
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections