20xx行人再識(shí)別年度進(jìn)展回顧-valse-資料下載頁(yè)
2025-08-05 18:38本頁(yè)面
【正文】 the generalization performance of reID models PersonX (Sun et al., CVPR 2022) ? How do viewpoint distributions in the Training Set affect model learning? ? How does Query viewpoint affect retrieval? ? How do true match viewpoints in the Gallery affect retrieval?v ? Eligibility ? Purity ? Sensitivity ? The controllability and quantification of synthetic training data can help to understanding the fundamental problems in person reID DGNet (Zheng et al., CVPR 2022) ? The controllability and quantification of synthetic training data can help to understanding the fundamental problems in person reID Outline ? Train/test on the same domain ? Unsupervised domain adaptation ? Person search ? Other problems ? Future research questions Future research questions ? Is largescale reid really solved? What the performance will be if we scale up the gallery to 1 million or 10 million images? How to accelerate? ? In unsupervised domain adaptation, person reid is a openset problem. What is its relationship with UDA in image classification? ? What is the relationship between person detection and reid? Can detection benefit reid, or the other way around? ? What are the differences between tracking and reid? Is it optimal to use reid features in tracking? ? How do environmental factors affect reid? Thank you!
環(huán)評(píng)公示相關(guān)推薦
鄂ICP備17016276號(hào)-1