English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Video object detection is challenging in the presence of appearancedeterioration in certain video frames. Therefore, it is a natural choice toaggregate temporal information from other frames of the same video into thecurrent frame. However, RoI Align, as one of the most core procedures of videodetectors, still remains extracting features from a single-frame feature mapfor proposals, making the extracted RoI features lack temporal information fromvideos. In this work, considering the features of the same object instance arehighly similar among frames in a video, a novel Temporal RoI Align operator isproposed to extract features from other frames feature maps for current frameproposals by utilizing feature similarity. The proposed Temporal RoI Alignoperator can extract temporal information from the entire video for proposals.We integrate it into single-frame video detectors and other state-of-the-artvideo detectors, and conduct quantitative experiments to demonstrate that theproposed Temporal RoI Align operator can consistently and significantly boostthe performance. Besides, the proposed Temporal RoI Align can also be appliedinto video instance segmentation. Codes are available athttps://github.com/open-mmlab/mmtracking

Temporal RoI Align for Video Object Recognition