Matches in SemOpenAlex for { <https://semopenalex.org/work/W2554247908> ?p ?o ?g. }
Showing items 1 to 84 of
84
with 100 items per page.
- W2554247908 abstract "This paper addresses the challenge of 3D human pose estimation from a single color image. Despite the general success of the end-to-end learning paradigm, top performing approaches employ a two-step solution consisting of a Convolutional Network (ConvNet) for 2D joint localization and a subsequent optimization step to recover 3D pose. In this paper, we identify the representation of 3D pose as a critical issue with current ConvNet approaches and make two important contributions towards validating the value of end-to-end learning for this task. First, we propose a fine discretization of the 3D space around the subject and train a ConvNet to predict per voxel likelihoods for each joint. This creates a natural representation for 3D pose and greatly improves performance over the direct regression of joint coordinates. Second, to further improve upon initial estimates, we employ a coarse-to-fine prediction scheme. This step addresses the large dimensionality increase and enables iterative refinement and repeated processing of the image features. The proposed approach outperforms all state-of-the-art methods on standard benchmarks achieving a relative error reduction greater than 30% on average. Additionally, we investigate using our volumetric representation in a related architecture which is suboptimal compared to our end-to-end approach, but is of practical interest, since it enables training when no image with corresponding 3D groundtruth is available, and allows us to present compelling results for in-the-wild images." @default.
- W2554247908 created "2016-11-30" @default.
- W2554247908 creator A5048438237 @default.
- W2554247908 creator A5050660826 @default.
- W2554247908 creator A5052269532 @default.
- W2554247908 creator A5060815622 @default.
- W2554247908 date "2017-07-01" @default.
- W2554247908 modified "2023-10-12" @default.
- W2554247908 title "Coarse-to-Fine Volumetric Prediction for Single-Image 3D Human Pose" @default.
- W2554247908 cites W1508437923 @default.
- W2554247908 cites W1905368000 @default.
- W2554247908 cites W1956126447 @default.
- W2554247908 cites W1997500560 @default.
- W2554247908 cites W2039262381 @default.
- W2554247908 cites W2054820429 @default.
- W2554247908 cites W2071882725 @default.
- W2554247908 cites W2079846689 @default.
- W2554247908 cites W2080873731 @default.
- W2554247908 cites W2088196373 @default.
- W2554247908 cites W2097117768 @default.
- W2554247908 cites W2097412577 @default.
- W2554247908 cites W2099333815 @default.
- W2554247908 cites W2101032778 @default.
- W2554247908 cites W2111446867 @default.
- W2554247908 cites W2123503110 @default.
- W2554247908 cites W2134704262 @default.
- W2554247908 cites W2135826343 @default.
- W2554247908 cites W2169738563 @default.
- W2554247908 cites W2171125807 @default.
- W2554247908 cites W2515603221 @default.
- W2554247908 cites W2520324844 @default.
- W2554247908 cites W2962729993 @default.
- W2554247908 cites W2963013806 @default.
- W2554247908 cites W2963474899 @default.
- W2554247908 cites W2963592930 @default.
- W2554247908 cites W2963688992 @default.
- W2554247908 cites W2963772981 @default.
- W2554247908 cites W2964225242 @default.
- W2554247908 cites W2964304707 @default.
- W2554247908 cites W602397586 @default.
- W2554247908 doi "https://doi.org/10.1109/cvpr.2017.139" @default.
- W2554247908 hasPublicationYear "2017" @default.
- W2554247908 type Work @default.
- W2554247908 sameAs 2554247908 @default.
- W2554247908 citedByCount "610" @default.
- W2554247908 countsByYear W25542479082017 @default.
- W2554247908 countsByYear W25542479082018 @default.
- W2554247908 countsByYear W25542479082019 @default.
- W2554247908 countsByYear W25542479082020 @default.
- W2554247908 countsByYear W25542479082021 @default.
- W2554247908 countsByYear W25542479082022 @default.
- W2554247908 countsByYear W25542479082023 @default.
- W2554247908 crossrefType "proceedings-article" @default.
- W2554247908 hasAuthorship W2554247908A5048438237 @default.
- W2554247908 hasAuthorship W2554247908A5050660826 @default.
- W2554247908 hasAuthorship W2554247908A5052269532 @default.
- W2554247908 hasAuthorship W2554247908A5060815622 @default.
- W2554247908 hasBestOaLocation W25542479082 @default.
- W2554247908 hasConcept C115961682 @default.
- W2554247908 hasConcept C154945302 @default.
- W2554247908 hasConcept C31972630 @default.
- W2554247908 hasConcept C41008148 @default.
- W2554247908 hasConceptScore W2554247908C115961682 @default.
- W2554247908 hasConceptScore W2554247908C154945302 @default.
- W2554247908 hasConceptScore W2554247908C31972630 @default.
- W2554247908 hasConceptScore W2554247908C41008148 @default.
- W2554247908 hasLocation W25542479081 @default.
- W2554247908 hasLocation W25542479082 @default.
- W2554247908 hasOpenAccess W2554247908 @default.
- W2554247908 hasPrimaryLocation W25542479081 @default.
- W2554247908 hasRelatedWork W2005185696 @default.
- W2554247908 hasRelatedWork W2092957489 @default.
- W2554247908 hasRelatedWork W2130228941 @default.
- W2554247908 hasRelatedWork W2132132164 @default.
- W2554247908 hasRelatedWork W2161229648 @default.
- W2554247908 hasRelatedWork W2235753890 @default.
- W2554247908 hasRelatedWork W2314419244 @default.
- W2554247908 hasRelatedWork W2366116130 @default.
- W2554247908 hasRelatedWork W2889893736 @default.
- W2554247908 hasRelatedWork W2993674027 @default.
- W2554247908 isParatext "false" @default.
- W2554247908 isRetracted "false" @default.
- W2554247908 magId "2554247908" @default.
- W2554247908 workType "article" @default.