Zobrazeno 1 - 10
of 15
pro vyhledávání: '"Amir Ghodrati"'
Publikováno v:
Lecture Notes in Computer Science ISBN: 9783031200793
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::a0993a01ebf4f7ec094495ed027b11d1
https://doi.org/10.1007/978-3-031-20080-9_18
https://doi.org/10.1007/978-3-031-20080-9_18
Publikováno v:
CVPR
In this paper, we propose a conditional early exiting framework for efficient video recognition. While existing works focus on selecting a subset of salient frames to reduce the computation costs, we propose to use a simple sampling strategy combined
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::3f1698ca9577c94e905d8995df4e5011
http://arxiv.org/abs/2104.13400
http://arxiv.org/abs/2104.13400
Publikováno v:
2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition: proceedings : virtual, 14-19 June 2020, 1168-1177
STARTPAGE=1168;ENDPAGE=1177;TITLE=2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition
CVPR
STARTPAGE=1168;ENDPAGE=1177;TITLE=2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition
CVPR
This paper tackles the problem of localizing actions in long untrimmed videos. Different from existing works, which all use annotated untrimmed videos during training, we learn only from short trimmed videos. This enables learning from large-scale da
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::d62dbda85e55a207ede9d6df46f2b1f4
https://dare.uva.nl/personal/pure/en/publications/actionbytes-learning-from-trimmed-videos-to-localize-actions(b8caedbc-5b42-4c10-9768-120170bcb6de).html
https://dare.uva.nl/personal/pure/en/publications/actionbytes-learning-from-trimmed-videos-to-localize-actions(b8caedbc-5b42-4c10-9768-120170bcb6de).html
Publikováno v:
International Journal of Computer Vision, 124(2), 115-131. Springer Netherlands
In this paper, a new method for generating object and action proposals in images and videos is proposed. It builds on activations of different convolutional layers of a pretrained CNN, combining the localization accuracy of the early layers with the
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c028726e0f7509e4509661ab6ec1085e
https://dare.uva.nl/personal/pure/en/publications/deepproposals-hunting-objects-and-actions-by-cascading-deep-convolutional-layers(ca77f056-9ffa-4407-a9c7-b85618c7c445).html
https://dare.uva.nl/personal/pure/en/publications/deepproposals-hunting-objects-and-actions-by-cascading-deep-convolutional-layers(ca77f056-9ffa-4407-a9c7-b85618c7c445).html
Publikováno v:
IEEE transactions on pattern analysis and machine intelligence
IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(4), 773-787. IEEE
IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(4), 773-787. IEEE
We propose a function-based temporal pooling method that captures the latent structure of the video sequence data - e.g. how frame-level features evolve over time in a video. We show how the parameters of a function that has been fit to the video dat
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::d3c62257ba07d71dfb26acab322f5a60
https://doi.org/10.1109/tpami.2016.2558148
https://doi.org/10.1109/tpami.2016.2558148
Publikováno v:
ICCV Workshops
This paper studies the effect of quality degradation, caused by lossy video compression, on video recognition. We investigate how the state of the art video enhancement restores the video quality needed for an effective video recognition. Furthermore
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::350ae08fb554246b0bac89284ba0d21d
https://doi.org/10.1109/iccvw.2019.00129
https://doi.org/10.1109/iccvw.2019.00129
Publikováno v:
2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition: proceedings : 18-22 June 2018, Salt Lake City, Utah, 5958-5966
STARTPAGE=5958;ENDPAGE=5966;TITLE=2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
CVPR
STARTPAGE=5958;ENDPAGE=5966;TITLE=2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
CVPR
This paper strives for pixel-level segmentation of actors and their actions in video content. Different from existing works, which all learn to segment from a fixed vocabulary of actor and action pairs, we infer the segmentation from a natural langua
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::2b3f65a2d9b3a8b108e8b14c449dadca
https://dare.uva.nl/personal/pure/en/publications/actor-and-action-video-segmentation-from-a-sentence(9e9e89b9-35e8-47f4-8e75-f46311eb7536).html
https://dare.uva.nl/personal/pure/en/publications/actor-and-action-video-segmentation-from-a-sentence(9e9e89b9-35e8-47f4-8e75-f46311eb7536).html
Publikováno v:
Lecture Notes in Computer Science ISBN: 9783319464534
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::d0417a47f347729d2fa6ad7e8eed492b
https://doi.org/10.1007/978-3-319-46454-1
https://doi.org/10.1007/978-3-319-46454-1
Autor:
Shohreh Kasaei, Amir Ghodrati
Publikováno v:
Intelligent Data Analysis. 16:537-550
New methods based on local spatio-temporal features have exhibited significant performance in action recognition. In these methods, feature selection plays an important role to achieve a superior performance. Actions are represented by local spatio-t
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::4e11b997415de29af4608d25fbc649e9
https://doi.org/10.3233/ida-2012-0538
https://doi.org/10.3233/ida-2012-0538
Publikováno v:
ICCV
In this paper we evaluate the quality of the activation layers of a convolutional neural network (CNN) for the gen- eration of object proposals. We generate hypotheses in a sliding-window fashion over different activation layers and show that the fin
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::773c3f664770cb0563c4318135edf398
https://doi.org/10.1109/iccv.2015.296
https://doi.org/10.1109/iccv.2015.296