FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3086333990> ?p ?o ?g. }

• W3086333990 abstract "One significant factor we expect the video representation learning to capture, especially in contrast with the image representation learning, is the object motion. However, we found that in the current mainstream video datasets, some action categories are highly related with the scene where the action happens, making the model tend to degrade to a solution where only the scene information is encoded. For example, a trained model may predict a video as playing football simply because it sees the field, neglecting that the subject is dancing as a cheerleader on the field. This is against our original intention towards the video representation learning and may bring scene bias on different dataset that can not be ignored. In order to tackle this problem, we propose to decouple the scene and the motion (DSM) with two simple operations, so that the model attention towards the motion information is better paid. Specifically, we construct a positive clip and a negative clip for each video. Compared to the original video, the positive/negative is motion-untouched/broken but scene-broken/untouched by Spatial Local Disturbance and Temporal Local Disturbance. Our objective is to pull the positive closer while pushing the negative farther to the original clip in the latent space. In this way, the impact of the scene is weakened while the temporal sensitivity of the network is further enhanced. We conduct experiments on two tasks with various backbones and different pre-training datasets, and find that our method surpass the SOTA methods with a remarkable 8.1% and 8.8% improvement towards action recognition task on the UCF101 and HMDB51 datasets respectively using the same backbone." @default.
• W3086333990 created "2020-09-21" @default.
• W3086333990 creator A5004402130 @default.
• W3086333990 creator A5011424573 @default.
• W3086333990 creator A5016080094 @default.
• W3086333990 creator A5017100408 @default.
• W3086333990 creator A5035927942 @default.
• W3086333990 creator A5041742366 @default.
• W3086333990 creator A5078102365 @default.
• W3086333990 creator A5082173143 @default.
• W3086333990 date "2020-09-12" @default.
• W3086333990 modified "2023-10-14" @default.
• W3086333990 title "Enhancing Unsupervised Video Representation Learning by Decoupling the Scene and the Motion" @default.
• W3086333990 cites W1522734439 @default.
• W3086333990 cites W2126579184 @default.
• W3086333990 cites W2138621090 @default.
• W3086333990 cites W2156303437 @default.
• W3086333990 cites W2295107390 @default.
• W3086333990 cites W2321533354 @default.
• W3086333990 cites W24089286 @default.
• W3086333990 cites W2487442924 @default.
• W3086333990 cites W2619947201 @default.
• W3086333990 cites W2751023760 @default.
• W3086333990 cites W2769112066 @default.
• W3086333990 cites W2770804203 @default.
• W3086333990 cites W2798271879 @default.
• W3086333990 cites W2798711578 @default.
• W3086333990 cites W2798991696 @default.
• W3086333990 cites W2799087757 @default.
• W3086333990 cites W2799146007 @default.
• W3086333990 cites W2895243423 @default.
• W3086333990 cites W2948242301 @default.
• W3086333990 cites W2951873722 @default.
• W3086333990 cites W2962742544 @default.
• W3086333990 cites W2962934715 @default.
• W3086333990 cites W2963315828 @default.
• W3086333990 cites W2963517393 @default.
• W3086333990 cites W2963524571 @default.
• W3086333990 cites W2963631366 @default.
• W3086333990 cites W2963814513 @default.
• W3086333990 cites W2964037671 @default.
• W3086333990 cites W2980139307 @default.
• W3086333990 cites W2990503944 @default.
• W3086333990 cites W2991133498 @default.
• W3086333990 cites W2995849700 @default.
• W3086333990 cites W2997907976 @default.
• W3086333990 cites W3005680577 @default.
• W3086333990 cites W3010874390 @default.
• W3086333990 cites W3034215340 @default.
• W3086333990 cites W3034381931 @default.
• W3086333990 cites W3034559532 @default.
• W3086333990 cites W3035524453 @default.
• W3086333990 cites W3047425522 @default.
• W3086333990 cites W3047826509 @default.
• W3086333990 cites W603908379 @default.
• W3086333990 doi "https://doi.org/10.48550/arxiv.2009.05757" @default.
• W3086333990 hasPublicationYear "2020" @default.
• W3086333990 type Work @default.
• W3086333990 sameAs 3086333990 @default.
• W3086333990 citedByCount "5" @default.
• W3086333990 countsByYear W30863339902021 @default.
• W3086333990 crossrefType "posted-content" @default.
• W3086333990 hasAuthorship W3086333990A5004402130 @default.
• W3086333990 hasAuthorship W3086333990A5011424573 @default.
• W3086333990 hasAuthorship W3086333990A5016080094 @default.
• W3086333990 hasAuthorship W3086333990A5017100408 @default.
• W3086333990 hasAuthorship W3086333990A5035927942 @default.
• W3086333990 hasAuthorship W3086333990A5041742366 @default.
• W3086333990 hasAuthorship W3086333990A5078102365 @default.
• W3086333990 hasAuthorship W3086333990A5082173143 @default.
• W3086333990 hasBestOaLocation W30863339901 @default.
• W3086333990 hasConcept C104114177 @default.
• W3086333990 hasConcept C121332964 @default.
• W3086333990 hasConcept C127413603 @default.
• W3086333990 hasConcept C133731056 @default.
• W3086333990 hasConcept C154945302 @default.
• W3086333990 hasConcept C17744445 @default.
• W3086333990 hasConcept C199360897 @default.
• W3086333990 hasConcept C199539241 @default.
• W3086333990 hasConcept C202444582 @default.
• W3086333990 hasConcept C205606062 @default.
• W3086333990 hasConcept C2776359362 @default.
• W3086333990 hasConcept C2780791683 @default.
• W3086333990 hasConcept C2780801425 @default.
• W3086333990 hasConcept C2781238097 @default.
• W3086333990 hasConcept C31972630 @default.
• W3086333990 hasConcept C33923547 @default.
• W3086333990 hasConcept C41008148 @default.
• W3086333990 hasConcept C59404180 @default.
• W3086333990 hasConcept C62520636 @default.
• W3086333990 hasConcept C94625758 @default.
• W3086333990 hasConcept C9652623 @default.
• W3086333990 hasConceptScore W3086333990C104114177 @default.
• W3086333990 hasConceptScore W3086333990C121332964 @default.
• W3086333990 hasConceptScore W3086333990C127413603 @default.
• W3086333990 hasConceptScore W3086333990C133731056 @default.
• W3086333990 hasConceptScore W3086333990C154945302 @default.
• W3086333990 hasConceptScore W3086333990C17744445 @default.
• W3086333990 hasConceptScore W3086333990C199360897 @default.
• W3086333990 hasConceptScore W3086333990C199539241 @default.

• next