Zobrazeno 1 - 9
of 9
pro vyhledávání: '"Tete Xiao"'
We present Region Similarity Representation Learning (ReSim), a new approach to self-supervised representation learning for localization-based tasks such as object detection and segmentation. While existing work has largely focused on solely learning
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::89e8d0d359e76205904fc07d9c5e9808
http://arxiv.org/abs/2103.12902
http://arxiv.org/abs/2103.12902
Publikováno v:
ICCV
Objects are entities we act upon, where the functionality of an object is determined by how we interact with it. In this work we propose a Dual Attention Network model which reasons about human-object interactions. The dual-attentional framework weig
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::9ddbdb96d687a4f3209d5afd233270f5
http://arxiv.org/abs/1909.04743
http://arxiv.org/abs/1909.04743
Publikováno v:
CVPR
Human action is naturally compositional: humans can easily recognize and perform actions with objects that are different from those used in training demonstrations. In this paper, we study the compositionality of action by looking into the dynamics o
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::ab1b23801cae5182dca6550f102e7802
Publikováno v:
CVPR
Detecting individual pedestrians in a crowd remains a challenging problem since the pedestrians often gather together and occlude each other in real-world scenarios. In this paper, we first explore how a state-of-the-art pedestrian detector is harmed
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::8933684679e11cc18e39dfcfdfd45e47
https://doi.org/10.1109/cvpr.2018.00811
https://doi.org/10.1109/cvpr.2018.00811
Autor:
Tete Xiao, Bolei Zhou, Hang Zhao, Xavier Puig, Adela Barriuso, Antonio Torralba, Sanja Fidler
Publikováno v:
arXiv
Semantic understanding of visual scenes is one of the holy grails of computer vision. Despite efforts of the community in data collection, there are still few image datasets covering a wide range of scenes and object categories with pixel-wise annota
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::81b997cbd59a7a72bf3f8d2dad328108
https://hdl.handle.net/1721.1/125771
https://hdl.handle.net/1721.1/125771
Publikováno v:
Computer Vision – ECCV 2018 ISBN: 9783030012632
ECCV (14)
ECCV (14)
Modern CNN-based object detectors rely on bounding box regression and non-maximum suppression to localize objects. While the probabilities for class labels naturally reflect classification confidence, localization confidence is absent. This makes pro
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::3153b1ced9c4ca215df71ef27da03b24
https://doi.org/10.1007/978-3-030-01264-9_48
https://doi.org/10.1007/978-3-030-01264-9_48
Publikováno v:
Computer Vision – ECCV 2018 ISBN: 9783030012274
ECCV (5)
ECCV (5)
Humans recognize the visual world at multiple levels: we effortlessly categorize scenes and detect objects inside, while also identifying the textures and surfaces of the objects along with their different compositional parts. In this paper, we study
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::cd976e69c2ac55df099b09e2eb4ee3d6
https://doi.org/10.1007/978-3-030-01228-1_26
https://doi.org/10.1007/978-3-030-01228-1_26
Publikováno v:
CVPR
The development of object detection in the era of deep learning, from R-CNN [11], Fast/Faster R-CNN [10, 31] to recent Mask R-CNN [14] and RetinaNet [24], mainly come from novel network, new framework, or loss design. However, mini-batch size, a key
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e977b635612202e79fe99756dbf920a5
http://arxiv.org/abs/1711.07240
http://arxiv.org/abs/1711.07240
Publikováno v:
CVPR
Aggregating extra features has been considered as an effective approach to boost traditional pedestrian detection methods. However, there is still a lack of studies on whether and how CNN-based pedestrian detectors can benefit from these extra featur
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::a8d7724d0bcaa57eb08061617adc6b6f