YOLACT++ Better Real-Time Instance Segmentation
Open accessAbstract
We present a simple, fully-convolutional model for real-time ( fps) instance segmentation that achieves competitive results on MS COCO evaluated on a single Titan Xp, which is significantly faster than any previous state-of-the-art approach. Moreover, we obtain this result after training on only one GPU. We accomplish this by breaking instance segmentation into two parallel subtasks: (1) generating a set of prototype masks and (2) predicting per-instance mask coefficients. Then we produce instance masks by linearly combining the prototypes with the mask coefficients. We find that because this process doesn't depend on repooling, this approach produces very high-quality masks and exhibits temporal stability for free. Furthermore, we analyze the emergent behavior of our prototypes and show they learn to localize instances on their own in a translation variant manner, despite being fully-convolutional. We also propose Fast NMS, a drop-in 12 ms faster replacement for standard NMS that only has a marginal performance penalty. Finally, by incorporating deformable convolutions into the backbone network, optimizing the prediction head with better anchor scales and aspect ratios, and adding a novel fast mask re-scoring branch, our YOLACT++ model can achieve 34.1 mAP on MS COCO at 33.5 fps, which is fairly close to the state-of-the-art approaches while still running at real-time.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Event cameras are especially well-suited for situations involving a lot of motion and difficult lighting conditions.
“thanks to these properties, event cameras are especially well-suited for situations involving a lot of motion, and difficult lighting conditions [142], [143].”
From A Systematic Review of Drone Based Road Traffic Monitoring System (2022) · cites this paper as [142]
Event cameras provide high temporal precision, high dynamical coverage, and low data rates.
“Event cameras, which provide high temporal precision, high dynamical coverage, and low data rates, represent a paradigm shift from conventional framebased cameras.”
From A Systematic Review of Drone Based Road Traffic Monitoring System (2022) · cites this paper as [142]
Event-based solutions are invariant to absolute illumination levels and more robust to high latency and motion blur issues.
“Indeed, eventbased solutions are invariant to absolute illumination levels and more robust to high latency and motion blur issues as reported in [142].”
From A Systematic Review of Drone Based Road Traffic Monitoring System (2022) · cites this paper as [142]
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · see citing works
You Only Look Once: Unified, Real-Time Object Detection
Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.91 · Cited by 53,803 · see citing works
Feature Pyramid Networks for Object Detection
Lin, T.-Y., Dollar, P., Girshick, R., He, K., Hariharan, B., & Belongie, S. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.106 · Cited by 30,939 · see citing works
Focal Loss for Dense Object Detection
Lin, T.-Y., Goyal, P., Girshick, R., He, K., & Dollar, P. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.324 · Cited by 30,581 · see citing works
SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation
Badrinarayanan, V., Kendall, A., & Cipolla, R. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2644615 · Cited by 19,631 · see citing works
Deformable Convolutional Networks
Dai, J., Qi, H., Xiong, Y., Li, Y., Zhang, G., Hu, H., & Wei, Y. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.89 · Cited by 7,513 · see citing works
Video Google: a text retrieval approach to object matching in videos
Sivic, & Zisserman (2003) · Proceedings Ninth IEEE International Conference on Computer Vision · doi:10.1109/iccv.2003.1238663 · Cited by 6,367 · see citing works
Locality-constrained Linear Coding for image classification
Wang, J., Yang, J., Yu, K., Lv, F., Huang, T., & Gong, Y. (2010) · 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2010.5540018 · Cited by 3,100 · see citing works
Deformable ConvNets V2: More Deformable, Better Results
Zhu, X., Hu, H., Lin, S., & Dai, J. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00953 · Cited by 2,943 · see citing works
Representing and Recognizing the Visual Appearance of Materials Using Three-Dimensional Textons
Leung, T., & Malik, J. (2001) · International Journal of Computer Vision · doi:10.1023/a:1011126920638 · Cited by 1,498 · cited 1× · see citing works
“Classical representations include textons [41] and visual words [42], with advances made via sparsity and locality”
BackgroundInstance-aware Semantic Segmentation via Multi-task Network Cascades
Dai, J., He, K., & Sun, J. (2015) · arXiv · doi:10.48550/arxiv.1512.04412 · Cited by 1,273 · see citing works
Focal Loss for Dense Object Detection
Lin, T.-Y., Goyal, P., Girshick, R., He, K., & Dollár, P. (2017) · arXiv · doi:10.48550/arxiv.1708.02002 · Cited by 1,227 · see citing works
Computing receptive fields of convolutional neural networks
Araujo, A., Norris, W., & Sim, J. (2019) · Distill · doi:10.23915/distill.00021 · Cited by 312 · cited 1× · see citing works
“For instance, ResNet 101 and ResNet 50 have 511px and 239px of padding in each direction respectively [50].”
BackgroundSegmentation-Aware Convolutional Networks Using Local Attention Masks
Harley, A. W., Derpanis, K. G., & Kokkinos, I. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.539 · Cited by 142 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
A Systematic Review of Drone Based Road Traffic Monitoring System
Bisio, I., Garibotto, C., Haleem, H., Lavagetto, F., & Sciarrone, A. (2022) · IEEE Access · doi:10.1109/access.2022.3207282 · Cited by 121 · see citing works
“Therefore, the design of specific 1272 Yolact ++ [142], and Yolactedge [143] can also be used to From traffic monitoring perspective, the goal of MOT is to track all the vehicles in an aerial sequence.”
BackgroundLe, T.-N., Nguyen, H. H., Yamagishi, J., & Echizen, I. (2021) · 2021 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00996 · Cited by 97 · see citing works
- Authors
- Bolya, D., Zhou, C., Xiao, F., & Lee, Y. J.
- Affiliations
- Georgia Institute of Technology (US)
- University of California, Davis (US)
- Venue
- IEEE Transactions on Pattern Analysis and Machine Intelligence
- Published
- 2022-02-01
- Volume
- 44
- Issue
- 2
- Pages
- 1108-1121
- ISSN
- 0162-8828 (ISSN-L)
- 2160-9292
- 1939-3539
- OpenAlex
- W3047032303
- Cited by
- 544
- References
- 14
- Citations by year
- 2026: 43 · 2025: 97 · 2024: 96 · 2023: 126 · 2022: 96 · 2021: 73 · 2020: 13
- FWCI
- 29.93
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Transactions on Pattern Analysis and Machine Intelligence (publishedVersion)
- PubMed (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- ARO (YIP W911NF17-1-0410)
- National Science Foundation (IIS-1751206, IIS-1812850)
- AWS ML Research Award
- Google Cloud Platform research credits
- XSEDE (IRI180001)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Visual Attention and Saliency Detection
- Keywords
- YOLACT, non-maximum suppression, deformable convolution, temporal stability, translation invariance, single-GPU training
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
