NUS Computer Vision and Machine Learning Lab (@cvml_nus) 's Twitter Profile
NUS Computer Vision and Machine Learning Lab

@cvml_nus

Computer Vision and Machine Learning Group at the National University of Singapore

ID: 1498967308110229507

linkhttps://sites.google.com/comp.nus.edu.sg/cvml/ calendar_today02-03-2022 10:26:30

11 Tweet

172 Takipçi

48 Takip Edilen

NUS Computer Vision and Machine Learning Lab (@cvml_nus) 's Twitter Profile Photo

Check out our new #ICLR2022 paper which studies the update behaviour of heatmaps in integral regression for human pose estimation. Paper: openreview.net/pdf?id=vHVcB-a…

Check out our new #ICLR2022 paper which studies the update behaviour of heatmaps in integral regression for human pose estimation.

Paper: openreview.net/pdf?id=vHVcB-a…
NUS Computer Vision and Machine Learning Lab (@cvml_nus) 's Twitter Profile Photo

Check out our new #CVPR2022 paper on semi-supervised video object segmentation. We propose an efficient plug-and-play acceleration framework by exploiting the temporal redundancies in the video presented by the compressed bitstream. Project: kai422.github.io/CoVOS/

Check out our new #CVPR2022 paper on semi-supervised video object segmentation. We propose an efficient plug-and-play acceleration framework by exploiting the temporal redundancies in the video presented by the compressed bitstream.

Project: kai422.github.io/CoVOS/
NUS Computer Vision and Machine Learning Lab (@cvml_nus) 's Twitter Profile Photo

Check out our new #CVPR2022 paper on the video deblurring task. We propose a memory-based video deblurring network that supplements useful information from the memory bank to the deblurring backbone. Paper: arxiv.org/abs/2204.02977

Check out our new #CVPR2022 paper on the video deblurring task. We propose a memory-based video deblurring network that supplements useful information from the memory bank to the deblurring backbone. 

Paper: arxiv.org/abs/2204.02977
Angela Yao (@angelayao101) 's Twitter Profile Photo

Our #CVPR22 paper presents Assembly101, a new large-scale procedural activity dataset with 513 hours of video of people assembling and disassembling toy vehicles. assembly-101.github.io Come and chat with us at CVPR to find out more!

NUS Computer Vision and Machine Learning Lab (@cvml_nus) 's Twitter Profile Photo

Check out our new #CVPR2022 paper that generates temporally coherent UV coordinates for loose clothing. Our method is not constrained by human body outlines. It can be used to flexibly synthesize new looks and modified visual styles. Paper: arxiv.org/pdf/2204.03671…

Check out our new #CVPR2022 paper that generates temporally coherent UV coordinates for loose clothing. Our method is not constrained by human body outlines. It can be used to flexibly synthesize new looks and modified visual styles.

Paper: arxiv.org/pdf/2204.03671…