The Lovász-Softmax loss: A tractable surrogate for the optimization of the intersection-over-union measure in neural networks
| Autor: | Maxim Berman, Amal Rannen Triki, Matthew B. Blaschko |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2018 |
| Předmět: |
FOS: Computer and information sciences
Jaccard index PSI_MBL Artificial neural network business.industry Computer science Deep learning Computer Vision and Pattern Recognition (cs.CV) Computer Science - Computer Vision and Pattern Recognition ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Context (language use) Pattern recognition 02 engineering and technology Image segmentation Scale invariance 010502 geochemistry & geophysics 01 natural sciences Submodular set function Softmax function 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Segmentation Artificial intelligence business 0105 earth and related environmental sciences |
| Zdroj: | CVPR |
| Popis: | 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 Lov\'asz 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. Comment: Accepted as a conference paper at CVPR 2018 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|