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

In recent years, we have seen the performance of video-based personRe-Identification (ReID) methods have improved considerably. However, most ofthe work in this area has dealt with videos acquired by fixed cameras withwider field of view. Recently, widespread use of wearable cameras and recordingdevices such as cellphones have opened the door to interesting research infirst-person Point-of-view (POV) videos (egocentric videos). Nonetheless,analysis of such videos is challenging due to factors such as poor videoquality due to ego-motion, blurriness, severe changes in lighting conditionsand perspective distortions. To facilitate the research towards conqueringthese challenges, this paper contributes a new dataset called EgoReID. Thedataset is captured using 3 mobile cellphones with non-overlappingfield-of-view. It contains 900 IDs and around 10,200 tracks with a total of176,000 detections. The dataset also contains 12-sensor meta data e.g. cameraorientation pitch and rotation for each video.  In addition, we propose a new framework which takes advantage of both visualand sensor meta data to successfully perform Person ReID. We extend image-basedre-ID method employing human body parsing trained on ten datasets tovideo-based re-ID. In our method, first frame level local features areextracted for each semantic region, then 3D convolutions are applied to encodethe temporal information in each sequence of semantic regions. Additionally, weemploy sensor meta data to predict targets' next camera and their estimatedtime of arrival, which considerably improves our ReID performance as itsignificantly reduces our search space.

EgoReID Dataset: Person Re-identification in Videos Acquired by Mobile Devices with First-Person Point-of-View