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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Although the anchor-based detectors have taken a big step forward inpedestrian detection, the overall performance of algorithm still needs furtherimprovement for practical applications, \emph{e.g.}, a good trade-off betweenthe accuracy and efficiency. To this end, this paper proposes a series ofsystematic optimization strategies for the detection pipeline of one-stagedetector, forming a single shot anchor-based detector (SADet) for efficient andaccurate pedestrian detection, which includes three main improvements. Firstly,we optimize the sample generation process by assigning soft tags to the outliersamples to generate semi-positive samples with continuous tag value between $0$and $1$, which not only produces more valid samples, but also strengthens therobustness of the model. Secondly, a novel Center-$IoU$ loss is applied as anew regression loss for bounding box regression, which not only retains thegood characteristics of IoU loss, but also solves some defects of it. Thirdly,we also design Cosine-NMS for the postprocess of predicted bounding boxes, andfurther propose adaptive anchor matching to enable the model to adaptivelymatch the anchor boxes to full or visible bounding boxes according to thedegree of occlusion, making the NMS and anchor matching algorithms moresuitable for occluded pedestrian detection. Though structurally simple, itpresents state-of-the-art result and real-time speed of $20$ FPS forVGA-resolution images ($640 \times 480$) on challenging pedestrian detectionbenchmarks, i.e., CityPersons, Caltech, and human detection benchmarkCrowdHuman, leading to a new attractive pedestrian detector.

SADet: Learning An Efficient and Accurate Pedestrian Detector