Papers › The Lovász-Softmax loss: A tractable surrogate for the optimization of the...

The Lovász-Softmax loss: A tractable surrogate for the optimization of the intersection-over-union measure in neural networks

24 May 2017CVPR 2018 6arXiv:1705.08790archive 2025-07-28

Maxim Berman, Amal Rannen Triki, Matthew B. Blaschko

The Jaccard index, also referred to as the intersection-over-union score, is commonly employed in the evaluation of image segmentation results given its perceptual qualities, scale invariance - which lends appropriate relevance to small objects, and appropriate counting of false negatives, in comparison to per-pixel losses. We present a method for direct optimization of the mean intersection-over-union loss in neural networks, in the context of semantic image segmentation, based on the convex Lovász extension of submodular losses. The loss is shown to perform better with respect to the Jaccard index measure than the traditionally used cross-entropy loss. We show quantitative and qualitative differences between optimizing the Jaccard index per image versus optimizing the Jaccard index taken over an entire dataset. We evaluate the impact of our method in a semantic segmentation pipeline and show substantially improved intersection-over-union segmentation scores on the Pascal VOC and Cityscapes datasets using state-of-the-art deep learning segmentation architectures.

PaperPDFCode

Syntology This paper's arXiv id is not in Syntology's Atlas id list (549,449 ids, as this build read them; the read time is in the build record), so this page links to no Syntology page for it. The list is not the whole graph; Syntology's feed shows papers as they enter the graph.

Code

bermanmaxim/LovaszSoftmax officialmentioned in papermentioned on GitHubtfSyntology: not harvested report
Gjiangtao/A-Deep-Supervised-Edge-Optimization-Algorithm-for-Salt-Body-Segmentation mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to SGDR: Stochastic Gradient Descent with Warm Restarts (arXiv:1608.03983; that paper's own run record: 16 ran (of which 7 constructed an object rather than computing a result; 14 with no instrument failure: 0 honoured, 0 violated, 14 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
adityajn105/brain-tumor-segmentation-unet mentioned on GitHubMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to U-Net: Convolutional Networks for Biomedical Image Segmentation (arXiv:1505.04597; that paper's own run record: 530 ran (of which 247 constructed an object rather than computing a result; 378 with no instrument failure: 18 honoured, 10 violated, 350 with no contract checked; 152 where Syntology's instrument failed) · 227 unverified); nothing checks that they implement this paper's method. report
CoinCheung/pytorch-loss pytorchMITSyntology: not harvested report

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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Image SegmentationSegmentationSemantic Segmentation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Real-Time Semantic Segmentation Cityscapes test ENet + Lovász-Softmax Frame (fps) 76.9 #36 of 39 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes test ENet + Lovász-Softmax Time (ms) 13 #36 of 39 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes test ENet + Lovász-Softmax mIoU 63.1% #36 of 39 Archive leaderboard report
Semantic Segmentation Cityscapes test ENet + Lovász-Softmax Mean IoU (class) 63.06% #99 of 105 Archive leaderboard report
Semantic Segmentation PASCAL VOC 2012 test Deeplab-v2 with Lovasz-Softmax loss Mean IoU 79.00% #34 of 51 Archive leaderboard report
Semantic Segmentation PASCAL VOC 2012 test Deeplab-v2 + Lovász-Softmax Mean IoU 79.0% #35 of 51 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.

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