Matches in SemOpenAlex for { <https://semopenalex.org/work/W3138275323> ?p ?o ?g. }
Showing items 1 to 79 of
79
with 100 items per page.
- W3138275323 abstract "In this work, we aim for temporally consistent semantic segmentation throughout frames in a video. Many semantic segmentation algorithms process images individually which leads to an inconsistent scene interpretation due to illumination changes, occlusions and other variations over time. To achieve a temporally consistent prediction, we train a convolutional neural network (CNN) which propagates features through consecutive frames in a video using a convolutional long short term memory (ConvLSTM) cell. Besides the temporal feature propagation, we penalize inconsistencies in our loss function. We show in our experiments that the performance improves when utilizing video information compared to single frame prediction. The mean intersection over union (mIoU) metric on the Cityscapes validation set increases from 45.2 % for the single frames to 57.9 % for video data after implementing the ConvLSTM to propagate features trough time on the ESPNet. Most importantly, inconsistency decreases from 4.5 % to 1.3 % which is a reduction by 71.1 %. Our results indicate that the added temporal information produces a frame-to-frame consistent and more accurate image understanding compared to single frame processing. Code and videos are available at this https URL" @default.
- W3138275323 created "2021-03-29" @default.
- W3138275323 creator A5060858108 @default.
- W3138275323 creator A5069159614 @default.
- W3138275323 date "2020-07-28" @default.
- W3138275323 modified "2023-09-28" @default.
- W3138275323 title "Frame-To-Frame Consistent Semantic Segmentation" @default.
- W3138275323 hasPublicationYear "2020" @default.
- W3138275323 type Work @default.
- W3138275323 sameAs 3138275323 @default.
- W3138275323 citedByCount "0" @default.
- W3138275323 crossrefType "journal-article" @default.
- W3138275323 hasAuthorship W3138275323A5060858108 @default.
- W3138275323 hasAuthorship W3138275323A5069159614 @default.
- W3138275323 hasConcept C126042441 @default.
- W3138275323 hasConcept C127413603 @default.
- W3138275323 hasConcept C138885662 @default.
- W3138275323 hasConcept C146978453 @default.
- W3138275323 hasConcept C153180895 @default.
- W3138275323 hasConcept C154945302 @default.
- W3138275323 hasConcept C162324750 @default.
- W3138275323 hasConcept C176217482 @default.
- W3138275323 hasConcept C177264268 @default.
- W3138275323 hasConcept C199360897 @default.
- W3138275323 hasConcept C21547014 @default.
- W3138275323 hasConcept C2776401178 @default.
- W3138275323 hasConcept C31972630 @default.
- W3138275323 hasConcept C41008148 @default.
- W3138275323 hasConcept C41895202 @default.
- W3138275323 hasConcept C58489278 @default.
- W3138275323 hasConcept C64543145 @default.
- W3138275323 hasConcept C76155785 @default.
- W3138275323 hasConcept C81363708 @default.
- W3138275323 hasConcept C89600930 @default.
- W3138275323 hasConceptScore W3138275323C126042441 @default.
- W3138275323 hasConceptScore W3138275323C127413603 @default.
- W3138275323 hasConceptScore W3138275323C138885662 @default.
- W3138275323 hasConceptScore W3138275323C146978453 @default.
- W3138275323 hasConceptScore W3138275323C153180895 @default.
- W3138275323 hasConceptScore W3138275323C154945302 @default.
- W3138275323 hasConceptScore W3138275323C162324750 @default.
- W3138275323 hasConceptScore W3138275323C176217482 @default.
- W3138275323 hasConceptScore W3138275323C177264268 @default.
- W3138275323 hasConceptScore W3138275323C199360897 @default.
- W3138275323 hasConceptScore W3138275323C21547014 @default.
- W3138275323 hasConceptScore W3138275323C2776401178 @default.
- W3138275323 hasConceptScore W3138275323C31972630 @default.
- W3138275323 hasConceptScore W3138275323C41008148 @default.
- W3138275323 hasConceptScore W3138275323C41895202 @default.
- W3138275323 hasConceptScore W3138275323C58489278 @default.
- W3138275323 hasConceptScore W3138275323C64543145 @default.
- W3138275323 hasConceptScore W3138275323C76155785 @default.
- W3138275323 hasConceptScore W3138275323C81363708 @default.
- W3138275323 hasConceptScore W3138275323C89600930 @default.
- W3138275323 hasOpenAccess W3138275323 @default.
- W3138275323 hasRelatedWork W1955404912 @default.
- W3138275323 hasRelatedWork W2175030374 @default.
- W3138275323 hasRelatedWork W2594448268 @default.
- W3138275323 hasRelatedWork W2894961607 @default.
- W3138275323 hasRelatedWork W2895012638 @default.
- W3138275323 hasRelatedWork W2923834406 @default.
- W3138275323 hasRelatedWork W2932883984 @default.
- W3138275323 hasRelatedWork W2951499547 @default.
- W3138275323 hasRelatedWork W2963268050 @default.
- W3138275323 hasRelatedWork W2964317599 @default.
- W3138275323 hasRelatedWork W3009478566 @default.
- W3138275323 hasRelatedWork W3009499989 @default.
- W3138275323 hasRelatedWork W3034657761 @default.
- W3138275323 hasRelatedWork W3035265644 @default.
- W3138275323 hasRelatedWork W3043117676 @default.
- W3138275323 hasRelatedWork W3046302739 @default.
- W3138275323 hasRelatedWork W3100270142 @default.
- W3138275323 hasRelatedWork W3133772107 @default.
- W3138275323 hasRelatedWork W3162395899 @default.
- W3138275323 hasRelatedWork W81520703 @default.
- W3138275323 isParatext "false" @default.
- W3138275323 isRetracted "false" @default.
- W3138275323 magId "3138275323" @default.
- W3138275323 workType "article" @default.