mlsanityData to 14 Oct 2026

Mask-aware IoU for Anchor Assignment in Real-time Instance Segmentation

arXiv 2110.09734 ↗First posted 19 Oct 2021cs.CVcited by 5

Not on the map. Why?

Abstract (arXiv)

This paper presents Mask-aware Intersection-over-Union (maIoU) for assigning anchor boxes as positives and negatives during training of instance segmentation methods. Unlike conventional IoU or its variants, which only considers the proximity of two boxes; maIoU consistently measures the proximity of an anchor box with not only a ground truth box but also its associated ground truth mask. Thus, additionally considering the mask, which, in fact, represents the shape of the object, maIoU enables a more accurate supervision during training. We present the effectiveness of maIoU on a state-of-the-art (SOTA) assigner, ATSS, by replacing IoU operation by our maIoU and training YOLACT, a SOTA real-time instance segmentation method. Using ATSS with maIoU consistently outperforms (i) ATSS with IoU by ~ 1 mask AP, (ii) baseline YOLACT with fixed IoU threshold assigner by ~ 2 mask AP over different image sizes and (iii) decreases the inference time by 25 % owing to using less anchors. Then, exploiting this efficiency, we devise maYOLACT, a faster and +6 AP more accurate detector than YOLACT. Our best model achieves 37.7 mask AP at 25 fps on COCO test-dev establishing a new state-of-the-art for real-time instance segmentation. Code is available at https://github.com/kemaloksuz/Mask-aware-IoU

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 35 of its 45 references that are in the map. First posted on arXiv 19 Oct 2021.

Builds on

35 of 45 references.

References · 35

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  3. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
  4. 2017Mask R-CNNcited by 7,277
  5. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 5,045
  6. 2016Feature Pyramid Networks for Object Detectioncited by 5,036
  7. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,444
  8. 2019FCOS: Fully Convolutional One-Stage Object Detectioncited by 1,577
  9. 2019MMDetection: Open MMLab Detection Toolbox and Benchmarkcited by 1,385
  10. 2019Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regressioncited by 1,368
  11. 2018Deformable ConvNets v2: More Deformable, Better Resultscited by 670
  12. 2019Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selectioncited by 540
23 more
  1. 2019Distance-IoU Loss: Faster and Better Learning for Bounding Box Regressioncited by 459
  2. 2019YOLACT: Real-time Instance Segmentationcited by 443
  3. 2019SOLO: Segmenting Objects by Locationscited by 289
  4. 2019Mask Scoring R-CNNcited by 266
  5. 2019PolarMask: Single Shot Instance Segmentation with Polar Representationcited by 188
  6. 2021Boundary IoU: Improving Object-Centric Image Segmentation Evaluationcited by 188
  7. 2020Probabilistic Anchor Assignment with IoU Prediction for Object Detectioncited by 150
  8. 2019CARAFE: Content-Aware ReAssembly of FEaturescited by 119
  9. 2019TensorMask: A Foundation for Dense Object Segmentationcited by 101
  10. 2020Enhancing Geometric Factors in Model Learning and Inference for Object Detection and Instance Segmentationcited by 98
  11. 2019YOLACT++: Better Real-time Instance Segmentationcited by 84
  12. 2019Learning from Noisy Anchors for One-stage Object Detectioncited by 35
  13. 2018Localization Recall Precision (LRP): A New Performance Metric for Object Detectioncited by 31
  14. 2020CenterMask: single shot instance segmentation with point representationcited by 31
  15. 2019Multiple Anchor Learning for Visual Object Detectioncited by 29
  16. 2021PolarMask++: Enhanced Polar Representation for Single-Shot Instance Segmentation and Beyondcited by 27
  17. 2019RetinaMask: Learning to predict masks improves state-of-the-art single-shot detection for freecited by 24
  18. 2021Rank & Sort Loss for Object Detection and Instance Segmentationcited by 22
  19. 2018MetaAnchor: Learning to Detect Objects with Customized Anchorscited by 21
  20. 2020One Metric to Measure them All: Localisation Recall Precision (LRP) for Evaluating Visual Detection Taskscited by 15
  21. 2018Anchor Box Optimization for Object Detectioncited by 8
  22. 2020Super-BPD: Super Boundary-to-Pixel Direction for Fast Image Segmentationcited by 8
  23. 2020AABO: Adaptive Anchor Box Optimization for Object Detection via Bayesian Sub-samplingcited by 5

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 3; 2024: 1.

5 papers.

  1. 2022Sparse Instance Activation for Real-Time Instance Segmentationcited by 63
  2. 2023MoCaE: Mixture of Calibrated Experts Significantly Improves Object Detectioncited by 17
  3. 2024Bucketed Ranking-based Losses for Efficient Training of Object Detectorscited by 3
  4. 2023Real-time Automatic M-mode Echocardiography Measurement with Panel Attention from Local-to-Global Pixelscited by 1
  5. 2023Generalized Mask-aware IoU for Anchor Assignment for Real-time Instance Segmentationcited by 0