Matches in SemOpenAlex for { <https://semopenalex.org/work/W3000338002> ?p ?o ?g. }
Showing items 1 to 94 of
94
with 100 items per page.
- W3000338002 abstract "In this paper, we propose an end-to-end self-supervised feature representation network for imitation learning. The proposed network incorporates a novel multi-level spatial attention module to amplify the relevant and suppress the irrelevant information while learning task-specific feature embeddings. The multi-level attention module takes multiple intermediate feature maps of the input image at different stages of the CNN pipeline and results a 2D matrix of compatibility scores for each feature map with respect to the given task. The weighted combination of the feature vectors with the scores estimated from attention modules leads to a more task specific feature representation of the input images. We thus name the proposed network as SMAK-Net, abbreviated from Self-supervised Multi-level spatial Attention Knowledge representation Network. We have trained this network using a metric learning loss which aims to decrease the distance between the feature representations of simultaneous frames from multiple view points and increases the distance between the neighboring frames of the same view point. The experiments are performed on the publicly available Multi-View pouring dataset [1]. The outputs of the attention module are demonstrated to highlight the task specific objects while suppressing the rest of the background in the input image. The proposed method is validated by qualitative and quantitative comparisons with the state-of-the art technique TCN [1] along with intensive ablation studies. This method is shown to significantly outperform TCN by 6.5% in the temporal alignment error metric while reducing the total number of training steps by 155K." @default.
- W3000338002 created "2020-01-23" @default.
- W3000338002 creator A5001713925 @default.
- W3000338002 creator A5003726743 @default.
- W3000338002 creator A5009996322 @default.
- W3000338002 creator A5047154762 @default.
- W3000338002 creator A5054693036 @default.
- W3000338002 date "2019-10-01" @default.
- W3000338002 modified "2023-10-11" @default.
- W3000338002 title "SMAK-Net: Self-Supervised Multi-level Spatial Attention Network for Knowledge Representation towards Imitation Learning" @default.
- W3000338002 cites W1836533770 @default.
- W3000338002 cites W1955055330 @default.
- W3000338002 cites W2095705004 @default.
- W3000338002 cites W2146444479 @default.
- W3000338002 cites W2172806452 @default.
- W3000338002 cites W2183341477 @default.
- W3000338002 cites W219040644 @default.
- W3000338002 cites W2302255633 @default.
- W3000338002 cites W2321739425 @default.
- W3000338002 cites W2475287302 @default.
- W3000338002 cites W2550462002 @default.
- W3000338002 cites W2555897561 @default.
- W3000338002 cites W2575671312 @default.
- W3000338002 cites W2769112066 @default.
- W3000338002 cites W2787420051 @default.
- W3000338002 cites W2897990074 @default.
- W3000338002 cites W2962756039 @default.
- W3000338002 cites W2963495494 @default.
- W3000338002 cites W2963715038 @default.
- W3000338002 cites W2964121744 @default.
- W3000338002 cites W2964345931 @default.
- W3000338002 cites W343636949 @default.
- W3000338002 doi "https://doi.org/10.1109/ro-man46459.2019.8956303" @default.
- W3000338002 hasPublicationYear "2019" @default.
- W3000338002 type Work @default.
- W3000338002 sameAs 3000338002 @default.
- W3000338002 citedByCount "0" @default.
- W3000338002 crossrefType "proceedings-article" @default.
- W3000338002 hasAuthorship W3000338002A5001713925 @default.
- W3000338002 hasAuthorship W3000338002A5003726743 @default.
- W3000338002 hasAuthorship W3000338002A5009996322 @default.
- W3000338002 hasAuthorship W3000338002A5047154762 @default.
- W3000338002 hasAuthorship W3000338002A5054693036 @default.
- W3000338002 hasConcept C119857082 @default.
- W3000338002 hasConcept C138885662 @default.
- W3000338002 hasConcept C153180895 @default.
- W3000338002 hasConcept C154945302 @default.
- W3000338002 hasConcept C162324750 @default.
- W3000338002 hasConcept C176217482 @default.
- W3000338002 hasConcept C17744445 @default.
- W3000338002 hasConcept C187736073 @default.
- W3000338002 hasConcept C199539241 @default.
- W3000338002 hasConcept C21547014 @default.
- W3000338002 hasConcept C2776359362 @default.
- W3000338002 hasConcept C2776401178 @default.
- W3000338002 hasConcept C2780451532 @default.
- W3000338002 hasConcept C41008148 @default.
- W3000338002 hasConcept C41895202 @default.
- W3000338002 hasConcept C59404180 @default.
- W3000338002 hasConcept C94625758 @default.
- W3000338002 hasConceptScore W3000338002C119857082 @default.
- W3000338002 hasConceptScore W3000338002C138885662 @default.
- W3000338002 hasConceptScore W3000338002C153180895 @default.
- W3000338002 hasConceptScore W3000338002C154945302 @default.
- W3000338002 hasConceptScore W3000338002C162324750 @default.
- W3000338002 hasConceptScore W3000338002C176217482 @default.
- W3000338002 hasConceptScore W3000338002C17744445 @default.
- W3000338002 hasConceptScore W3000338002C187736073 @default.
- W3000338002 hasConceptScore W3000338002C199539241 @default.
- W3000338002 hasConceptScore W3000338002C21547014 @default.
- W3000338002 hasConceptScore W3000338002C2776359362 @default.
- W3000338002 hasConceptScore W3000338002C2776401178 @default.
- W3000338002 hasConceptScore W3000338002C2780451532 @default.
- W3000338002 hasConceptScore W3000338002C41008148 @default.
- W3000338002 hasConceptScore W3000338002C41895202 @default.
- W3000338002 hasConceptScore W3000338002C59404180 @default.
- W3000338002 hasConceptScore W3000338002C94625758 @default.
- W3000338002 hasLocation W30003380021 @default.
- W3000338002 hasOpenAccess W3000338002 @default.
- W3000338002 hasPrimaryLocation W30003380021 @default.
- W3000338002 hasRelatedWork W10202958 @default.
- W3000338002 hasRelatedWork W1243554 @default.
- W3000338002 hasRelatedWork W1383942 @default.
- W3000338002 hasRelatedWork W14516383 @default.
- W3000338002 hasRelatedWork W14789944 @default.
- W3000338002 hasRelatedWork W4085024 @default.
- W3000338002 hasRelatedWork W5834068 @default.
- W3000338002 hasRelatedWork W7382899 @default.
- W3000338002 hasRelatedWork W9770290 @default.
- W3000338002 hasRelatedWork W9860846 @default.
- W3000338002 isParatext "false" @default.
- W3000338002 isRetracted "false" @default.
- W3000338002 magId "3000338002" @default.
- W3000338002 workType "article" @default.