Matches in SemOpenAlex for { <https://semopenalex.org/work/W2955368974> ?p ?o ?g. }
- W2955368974 abstract "The study of object representations in computer vision has primarily focused on developing representations that are useful for image classification, object detection, or semantic segmentation as downstream tasks. In this work we aim to learn object representations that are useful for control and reinforcement learning (RL). To this end, we introduce Transporter, a neural network architecture for discovering concise geometric object representations in terms of keypoints or image-space coordinates. Our method learns from raw video frames in a fully unsupervised manner, by transporting learnt image features between video frames using a keypoint bottleneck. The discovered keypoints track objects and object parts across long time-horizons more accurately than recent similar methods. Furthermore, consistent long-term tracking enables two notable results in control domains -- (1) using the keypoint co-ordinates and corresponding image features as inputs enables highly sample-efficient reinforcement learning; (2) learning to explore by controlling keypoint locations drastically reduces the search space, enabling deep exploration (leading to states unreachable through random action exploration) without any extrinsic rewards." @default.
- W2955368974 created "2019-07-12" @default.
- W2955368974 creator A5008904142 @default.
- W2955368974 creator A5046449484 @default.
- W2955368974 creator A5057678172 @default.
- W2955368974 creator A5059460567 @default.
- W2955368974 creator A5086192896 @default.
- W2955368974 creator A5089497713 @default.
- W2955368974 creator A5091197128 @default.
- W2955368974 date "2019-06-19" @default.
- W2955368974 modified "2023-09-27" @default.
- W2955368974 title "Unsupervised Learning of Object Keypoints for Perception and Control" @default.
- W2955368974 cites W1522301498 @default.
- W2955368974 cites W166862392 @default.
- W2955368974 cites W1691728462 @default.
- W2955368974 cites W1903029394 @default.
- W2955368974 cites W2030346542 @default.
- W2955368974 cites W2099471712 @default.
- W2955368974 cites W2109910161 @default.
- W2955368974 cites W2145339207 @default.
- W2955368974 cites W2163605009 @default.
- W2955368974 cites W2257979135 @default.
- W2955368974 cites W2281112906 @default.
- W2955368974 cites W2322480645 @default.
- W2955368974 cites W2623491082 @default.
- W2955368974 cites W2736601468 @default.
- W2955368974 cites W2751973545 @default.
- W2955368974 cites W2753738274 @default.
- W2955368974 cites W2781585732 @default.
- W2955368974 cites W2786118190 @default.
- W2955368974 cites W2807725536 @default.
- W2955368974 cites W2812468425 @default.
- W2955368974 cites W2883594813 @default.
- W2955368974 cites W2887997593 @default.
- W2955368974 cites W2890967717 @default.
- W2955368974 cites W2911448865 @default.
- W2955368974 cites W2913407674 @default.
- W2955368974 cites W2914261249 @default.
- W2955368974 cites W2924027593 @default.
- W2955368974 cites W2935920407 @default.
- W2955368974 cites W2949117887 @default.
- W2955368974 cites W2962981304 @default.
- W2955368974 cites W2963022858 @default.
- W2955368974 cites W2963226019 @default.
- W2955368974 cites W2963262099 @default.
- W2955368974 cites W2963296584 @default.
- W2955368974 cites W2963419579 @default.
- W2955368974 cites W2963547393 @default.
- W2955368974 cites W2963634205 @default.
- W2955368974 cites W2963717490 @default.
- W2955368974 cites W2964191931 @default.
- W2955368974 cites W2964226882 @default.
- W2955368974 cites W2964291307 @default.
- W2955368974 hasPublicationYear "2019" @default.
- W2955368974 type Work @default.
- W2955368974 sameAs 2955368974 @default.
- W2955368974 citedByCount "12" @default.
- W2955368974 countsByYear W29553689742019 @default.
- W2955368974 countsByYear W29553689742020 @default.
- W2955368974 countsByYear W29553689742021 @default.
- W2955368974 crossrefType "posted-content" @default.
- W2955368974 hasAuthorship W2955368974A5008904142 @default.
- W2955368974 hasAuthorship W2955368974A5046449484 @default.
- W2955368974 hasAuthorship W2955368974A5057678172 @default.
- W2955368974 hasAuthorship W2955368974A5059460567 @default.
- W2955368974 hasAuthorship W2955368974A5086192896 @default.
- W2955368974 hasAuthorship W2955368974A5089497713 @default.
- W2955368974 hasAuthorship W2955368974A5091197128 @default.
- W2955368974 hasConcept C149635348 @default.
- W2955368974 hasConcept C153180895 @default.
- W2955368974 hasConcept C154945302 @default.
- W2955368974 hasConcept C2776151529 @default.
- W2955368974 hasConcept C2780513914 @default.
- W2955368974 hasConcept C2781238097 @default.
- W2955368974 hasConcept C31972630 @default.
- W2955368974 hasConcept C41008148 @default.
- W2955368974 hasConcept C50644808 @default.
- W2955368974 hasConcept C8038995 @default.
- W2955368974 hasConcept C89600930 @default.
- W2955368974 hasConcept C97541855 @default.
- W2955368974 hasConceptScore W2955368974C149635348 @default.
- W2955368974 hasConceptScore W2955368974C153180895 @default.
- W2955368974 hasConceptScore W2955368974C154945302 @default.
- W2955368974 hasConceptScore W2955368974C2776151529 @default.
- W2955368974 hasConceptScore W2955368974C2780513914 @default.
- W2955368974 hasConceptScore W2955368974C2781238097 @default.
- W2955368974 hasConceptScore W2955368974C31972630 @default.
- W2955368974 hasConceptScore W2955368974C41008148 @default.
- W2955368974 hasConceptScore W2955368974C50644808 @default.
- W2955368974 hasConceptScore W2955368974C8038995 @default.
- W2955368974 hasConceptScore W2955368974C89600930 @default.
- W2955368974 hasConceptScore W2955368974C97541855 @default.
- W2955368974 hasLocation W29553689741 @default.
- W2955368974 hasOpenAccess W2955368974 @default.
- W2955368974 hasPrimaryLocation W29553689741 @default.
- W2955368974 hasRelatedWork W2016920098 @default.
- W2955368974 hasRelatedWork W2031688197 @default.
- W2955368974 hasRelatedWork W2237427287 @default.
- W2955368974 hasRelatedWork W2406445210 @default.
- W2955368974 hasRelatedWork W2611103765 @default.