Matches in SemOpenAlex for { <https://semopenalex.org/work/W3035557275> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W3035557275 abstract "This paper is about weakly supervised action segmentation, where the ground truth specifies only a set of actions present in a training video, but not their true temporal ordering. Prior work typically uses a classifier that independently labels video frames for generating the pseudo ground truth, and multiple instance learning for training the classifier. We extend this framework by specifying an HMM, which accounts for co-occurrences of action classes and their temporal lengths, and by explicitly training the HMM on a Viterbi-based loss. Our first contribution is the formulation of a new set-constrained Viterbi algorithm (SCV). Given a video, the SCV generates the MAP action segmentation that satisfies the ground truth. This prediction is used as a framewise pseudo ground truth in our HMM training. Our second contribution in training is a new regularization of feature affinities between training videos that share the same action classes. Evaluation on action segmentation and alignment on the Breakfast, MPII Cooking2, Hollywood Extended datasets demonstrates our significant performance improvement for the two tasks over prior work." @default.
- W3035557275 created "2020-06-19" @default.
- W3035557275 creator A5027742996 @default.
- W3035557275 creator A5027835055 @default.
- W3035557275 date "2020-06-01" @default.
- W3035557275 modified "2023-10-18" @default.
- W3035557275 title "Set-Constrained Viterbi for Set-Supervised Action Segmentation" @default.
- W3035557275 cites W2019660985 @default.
- W3035557275 cites W2033740597 @default.
- W3035557275 cites W2099614498 @default.
- W3035557275 cites W2105101328 @default.
- W3035557275 cites W2463640844 @default.
- W3035557275 cites W2530494944 @default.
- W3035557275 cites W2600081845 @default.
- W3035557275 cites W2604113307 @default.
- W3035557275 cites W2755802490 @default.
- W3035557275 cites W2798345491 @default.
- W3035557275 cites W2962916463 @default.
- W3035557275 cites W2963045696 @default.
- W3035557275 cites W2963642716 @default.
- W3035557275 cites W2975813532 @default.
- W3035557275 cites W3009622574 @default.
- W3035557275 cites W3123077146 @default.
- W3035557275 doi "https://doi.org/10.1109/cvpr42600.2020.01083" @default.
- W3035557275 hasPublicationYear "2020" @default.
- W3035557275 type Work @default.
- W3035557275 sameAs 3035557275 @default.
- W3035557275 citedByCount "27" @default.
- W3035557275 countsByYear W30355572752021 @default.
- W3035557275 countsByYear W30355572752022 @default.
- W3035557275 countsByYear W30355572752023 @default.
- W3035557275 crossrefType "proceedings-article" @default.
- W3035557275 hasAuthorship W3035557275A5027742996 @default.
- W3035557275 hasAuthorship W3035557275A5027835055 @default.
- W3035557275 hasBestOaLocation W30355572752 @default.
- W3035557275 hasConcept C11413529 @default.
- W3035557275 hasConcept C119857082 @default.
- W3035557275 hasConcept C124504099 @default.
- W3035557275 hasConcept C130319729 @default.
- W3035557275 hasConcept C146849305 @default.
- W3035557275 hasConcept C153180895 @default.
- W3035557275 hasConcept C154945302 @default.
- W3035557275 hasConcept C157125643 @default.
- W3035557275 hasConcept C193969084 @default.
- W3035557275 hasConcept C23224414 @default.
- W3035557275 hasConcept C41008148 @default.
- W3035557275 hasConcept C57273362 @default.
- W3035557275 hasConcept C60582962 @default.
- W3035557275 hasConcept C89600930 @default.
- W3035557275 hasConcept C95623464 @default.
- W3035557275 hasConceptScore W3035557275C11413529 @default.
- W3035557275 hasConceptScore W3035557275C119857082 @default.
- W3035557275 hasConceptScore W3035557275C124504099 @default.
- W3035557275 hasConceptScore W3035557275C130319729 @default.
- W3035557275 hasConceptScore W3035557275C146849305 @default.
- W3035557275 hasConceptScore W3035557275C153180895 @default.
- W3035557275 hasConceptScore W3035557275C154945302 @default.
- W3035557275 hasConceptScore W3035557275C157125643 @default.
- W3035557275 hasConceptScore W3035557275C193969084 @default.
- W3035557275 hasConceptScore W3035557275C23224414 @default.
- W3035557275 hasConceptScore W3035557275C41008148 @default.
- W3035557275 hasConceptScore W3035557275C57273362 @default.
- W3035557275 hasConceptScore W3035557275C60582962 @default.
- W3035557275 hasConceptScore W3035557275C89600930 @default.
- W3035557275 hasConceptScore W3035557275C95623464 @default.
- W3035557275 hasLocation W30355572751 @default.
- W3035557275 hasLocation W30355572752 @default.
- W3035557275 hasOpenAccess W3035557275 @default.
- W3035557275 hasPrimaryLocation W30355572751 @default.
- W3035557275 hasRelatedWork W1975869217 @default.
- W3035557275 hasRelatedWork W1987061770 @default.
- W3035557275 hasRelatedWork W2062804535 @default.
- W3035557275 hasRelatedWork W2116722627 @default.
- W3035557275 hasRelatedWork W2136652457 @default.
- W3035557275 hasRelatedWork W2143297499 @default.
- W3035557275 hasRelatedWork W2160171981 @default.
- W3035557275 hasRelatedWork W2169849734 @default.
- W3035557275 hasRelatedWork W2385954530 @default.
- W3035557275 hasRelatedWork W2993094530 @default.
- W3035557275 isParatext "false" @default.
- W3035557275 isRetracted "false" @default.
- W3035557275 magId "3035557275" @default.
- W3035557275 workType "article" @default.