Matches in SemOpenAlex for { <https://semopenalex.org/work/W3134150215> ?p ?o ?g. }
- W3134150215 abstract "Semantic segmentation is an extensively studied task in computer vision, with numerous methods proposed every year. Thanks to the advent of deep learning in semantic segmentation, the performance on existing benchmarks is close to saturation. A natural question then arises: Does the superior performance on the closed (and frequently re-used) test sets transfer to the open visual world with unconstrained variations? In this paper, we take steps toward answering the question by exposing failures of existing semantic segmentation methods in the open visual world under the constraint of very limited human labeling effort. Inspired by previous research on model falsification, we start from an arbitrarily large image set, and automatically sample a small image set by MAximizing the Discrepancy (MAD) between two segmentation methods. The selected images have the greatest potential in falsifying either (or both) of the two methods. We also explicitly enforce several conditions to diversify the exposed failures, corresponding to different underlying root causes. A segmentation method, whose failures are more difficult to be exposed in the MAD competition, is considered better. We conduct a thorough MAD diagnosis of ten PASCAL VOC semantic segmentation algorithms. With detailed analysis of experimental results, we point out strengths and weaknesses of the competing algorithms, as well as potential research directions for further advancement in semantic segmentation. The codes are publicly available at url{https://github.com/QTJiebin/MAD_Segmentation}." @default.
- W3134150215 created "2021-03-15" @default.
- W3134150215 creator A5020029652 @default.
- W3134150215 creator A5047172059 @default.
- W3134150215 creator A5048522863 @default.
- W3134150215 creator A5055987663 @default.
- W3134150215 creator A5063013411 @default.
- W3134150215 date "2021-02-27" @default.
- W3134150215 modified "2023-10-16" @default.
- W3134150215 title "Exposing Semantic Segmentation Failures via Maximum Discrepancy Competition" @default.
- W3134150215 cites W109452506 @default.
- W3134150215 cites W1565327149 @default.
- W3134150215 cites W1567215857 @default.
- W3134150215 cites W1686810756 @default.
- W3134150215 cites W1699734612 @default.
- W3134150215 cites W1901129140 @default.
- W3134150215 cites W1903029394 @default.
- W3134150215 cites W1945616565 @default.
- W3134150215 cites W2038721957 @default.
- W3134150215 cites W2083542343 @default.
- W3134150215 cites W2085518012 @default.
- W3134150215 cites W2095389390 @default.
- W3134150215 cites W2100588357 @default.
- W3134150215 cites W2108598243 @default.
- W3134150215 cites W2109812093 @default.
- W3134150215 cites W2112301665 @default.
- W3134150215 cites W2117539524 @default.
- W3134150215 cites W2126226185 @default.
- W3134150215 cites W2127006916 @default.
- W3134150215 cites W2133665775 @default.
- W3134150215 cites W2161907179 @default.
- W3134150215 cites W2161969291 @default.
- W3134150215 cites W2163371195 @default.
- W3134150215 cites W2163605009 @default.
- W3134150215 cites W2164877691 @default.
- W3134150215 cites W2168356304 @default.
- W3134150215 cites W2177274842 @default.
- W3134150215 cites W2194775991 @default.
- W3134150215 cites W2286929393 @default.
- W3134150215 cites W2340897893 @default.
- W3134150215 cites W2549139847 @default.
- W3134150215 cites W2560023338 @default.
- W3134150215 cites W2562510222 @default.
- W3134150215 cites W2592939477 @default.
- W3134150215 cites W2598666589 @default.
- W3134150215 cites W2613718673 @default.
- W3134150215 cites W2616028256 @default.
- W3134150215 cites W2630837129 @default.
- W3134150215 cites W2781228439 @default.
- W3134150215 cites W2804488433 @default.
- W3134150215 cites W2893890202 @default.
- W3134150215 cites W2895340641 @default.
- W3134150215 cites W2898726413 @default.
- W3134150215 cites W2905631704 @default.
- W3134150215 cites W2946824094 @default.
- W3134150215 cites W2952610664 @default.
- W3134150215 cites W2955058313 @default.
- W3134150215 cites W2962747122 @default.
- W3134150215 cites W2962772649 @default.
- W3134150215 cites W2962850830 @default.
- W3134150215 cites W2962900737 @default.
- W3134150215 cites W2963728677 @default.
- W3134150215 cites W2964309882 @default.
- W3134150215 cites W2979699022 @default.
- W3134150215 cites W2979997102 @default.
- W3134150215 cites W2981689412 @default.
- W3134150215 cites W2993235622 @default.
- W3134150215 cites W2995389272 @default.
- W3134150215 cites W2996048612 @default.
- W3134150215 cites W2996621846 @default.
- W3134150215 cites W3034355852 @default.
- W3134150215 cites W3034958977 @default.
- W3134150215 cites W3108490213 @default.
- W3134150215 cites W845365781 @default.
- W3134150215 doi "https://doi.org/10.48550/arxiv.2103.00259" @default.
- W3134150215 hasPublicationYear "2021" @default.
- W3134150215 type Work @default.
- W3134150215 sameAs 3134150215 @default.
- W3134150215 citedByCount "0" @default.
- W3134150215 crossrefType "posted-content" @default.
- W3134150215 hasAuthorship W3134150215A5020029652 @default.
- W3134150215 hasAuthorship W3134150215A5047172059 @default.
- W3134150215 hasAuthorship W3134150215A5048522863 @default.
- W3134150215 hasAuthorship W3134150215A5055987663 @default.
- W3134150215 hasAuthorship W3134150215A5063013411 @default.
- W3134150215 hasBestOaLocation W31341502151 @default.
- W3134150215 hasConcept C119857082 @default.
- W3134150215 hasConcept C124504099 @default.
- W3134150215 hasConcept C153180895 @default.
- W3134150215 hasConcept C154945302 @default.
- W3134150215 hasConcept C199360897 @default.
- W3134150215 hasConcept C41008148 @default.
- W3134150215 hasConcept C65885262 @default.
- W3134150215 hasConcept C75608658 @default.
- W3134150215 hasConcept C89600930 @default.
- W3134150215 hasConceptScore W3134150215C119857082 @default.
- W3134150215 hasConceptScore W3134150215C124504099 @default.
- W3134150215 hasConceptScore W3134150215C153180895 @default.
- W3134150215 hasConceptScore W3134150215C154945302 @default.
- W3134150215 hasConceptScore W3134150215C199360897 @default.