Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313328085> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W4313328085 endingPage "400" @default.
- W4313328085 startingPage "400" @default.
- W4313328085 abstract "Anticipation of per-pixel semantics in a future unobserved frame is also known as dense semantic forecasting. State-of-the-art methods are based on single-level regression of a subsampled abstract representation of a recognition model. However, single-level regression cannot account for skip connections from the backbone to the upsampling path. We propose to address this shortcoming by warping shallow features from observed images with upsampled feature flow. Our goal is not straightforward, since warping with coarse feature flow introduces noise into the forecasted features. We therefore base our work on single-frame models that are more resistant to the noise in skip connections. To achieve this, we propose a training procedure that enables recognition models to operate reasonably well with or without skip connections. Validation experiments reveal interesting insights into the influence of particular skip connections on recognition accuracy. Our forecasting method delivers 70.2% mIoU 0.18 s into the future and 58.5% mIoU 0.54 s into the future. These experiments show 0.6 mIoU points of improved accuracy with respect to the baseline and reveal promising directions for future work." @default.
- W4313328085 created "2023-01-06" @default.
- W4313328085 creator A5032164394 @default.
- W4313328085 creator A5073708578 @default.
- W4313328085 creator A5088679082 @default.
- W4313328085 date "2022-12-28" @default.
- W4313328085 modified "2023-09-26" @default.
- W4313328085 title "Dense Semantic Forecasting with Multi-Level Feature Warping" @default.
- W4313328085 cites W1901129140 @default.
- W4313328085 cites W1903029394 @default.
- W4313328085 cites W2108598243 @default.
- W4313328085 cites W2194775991 @default.
- W4313328085 cites W2340897893 @default.
- W4313328085 cites W2412782625 @default.
- W4313328085 cites W2422305492 @default.
- W4313328085 cites W2424778531 @default.
- W4313328085 cites W2560023338 @default.
- W4313328085 cites W2565639579 @default.
- W4313328085 cites W2601686579 @default.
- W4313328085 cites W2891859485 @default.
- W4313328085 cites W2963419596 @default.
- W4313328085 cites W2963446712 @default.
- W4313328085 cites W2963482775 @default.
- W4313328085 cites W2963881378 @default.
- W4313328085 cites W2964309882 @default.
- W4313328085 cites W2965191446 @default.
- W4313328085 cites W2968393007 @default.
- W4313328085 cites W2982562928 @default.
- W4313328085 cites W3016476472 @default.
- W4313328085 cites W3017672812 @default.
- W4313328085 cites W3023322676 @default.
- W4313328085 cites W3034355852 @default.
- W4313328085 cites W3094891242 @default.
- W4313328085 cites W3105636206 @default.
- W4313328085 cites W3149726007 @default.
- W4313328085 cites W3202749605 @default.
- W4313328085 cites W4200339473 @default.
- W4313328085 doi "https://doi.org/10.3390/app13010400" @default.
- W4313328085 hasPublicationYear "2022" @default.
- W4313328085 type Work @default.
- W4313328085 citedByCount "0" @default.
- W4313328085 crossrefType "journal-article" @default.
- W4313328085 hasAuthorship W4313328085A5032164394 @default.
- W4313328085 hasAuthorship W4313328085A5073708578 @default.
- W4313328085 hasAuthorship W4313328085A5088679082 @default.
- W4313328085 hasBestOaLocation W43133280851 @default.
- W4313328085 hasConcept C110384440 @default.
- W4313328085 hasConcept C115961682 @default.
- W4313328085 hasConcept C119857082 @default.
- W4313328085 hasConcept C138885662 @default.
- W4313328085 hasConcept C153180895 @default.
- W4313328085 hasConcept C154945302 @default.
- W4313328085 hasConcept C157202957 @default.
- W4313328085 hasConcept C184337299 @default.
- W4313328085 hasConcept C199360897 @default.
- W4313328085 hasConcept C2776401178 @default.
- W4313328085 hasConcept C41008148 @default.
- W4313328085 hasConcept C41895202 @default.
- W4313328085 hasConcept C88516994 @default.
- W4313328085 hasConcept C99498987 @default.
- W4313328085 hasConceptScore W4313328085C110384440 @default.
- W4313328085 hasConceptScore W4313328085C115961682 @default.
- W4313328085 hasConceptScore W4313328085C119857082 @default.
- W4313328085 hasConceptScore W4313328085C138885662 @default.
- W4313328085 hasConceptScore W4313328085C153180895 @default.
- W4313328085 hasConceptScore W4313328085C154945302 @default.
- W4313328085 hasConceptScore W4313328085C157202957 @default.
- W4313328085 hasConceptScore W4313328085C184337299 @default.
- W4313328085 hasConceptScore W4313328085C199360897 @default.
- W4313328085 hasConceptScore W4313328085C2776401178 @default.
- W4313328085 hasConceptScore W4313328085C41008148 @default.
- W4313328085 hasConceptScore W4313328085C41895202 @default.
- W4313328085 hasConceptScore W4313328085C88516994 @default.
- W4313328085 hasConceptScore W4313328085C99498987 @default.
- W4313328085 hasFunder F4320322674 @default.
- W4313328085 hasIssue "1" @default.
- W4313328085 hasLocation W43133280851 @default.
- W4313328085 hasOpenAccess W4313328085 @default.
- W4313328085 hasPrimaryLocation W43133280851 @default.
- W4313328085 hasRelatedWork W1991441988 @default.
- W4313328085 hasRelatedWork W2101823270 @default.
- W4313328085 hasRelatedWork W2116300362 @default.
- W4313328085 hasRelatedWork W2382607599 @default.
- W4313328085 hasRelatedWork W2546942002 @default.
- W4313328085 hasRelatedWork W2922188210 @default.
- W4313328085 hasRelatedWork W2970216048 @default.
- W4313328085 hasRelatedWork W3144145640 @default.
- W4313328085 hasRelatedWork W3171287051 @default.
- W4313328085 hasRelatedWork W1487047144 @default.
- W4313328085 hasVolume "13" @default.
- W4313328085 isParatext "false" @default.
- W4313328085 isRetracted "false" @default.
- W4313328085 workType "article" @default.