MI3C: Mining intra- and inter-image context for person search

Zongheng Tang, Yulu Gao, Tianrui Hui, Fengguang Peng, Si Liu

Abstract: Highlights•We propose a unified framework termed MI3<math><msup is="true"><mrow is="true"></mrow><mrow is="true"><mn is="true">3</mn></mrow></msup></math>C, which tackles person search from a more comprehensive perspective by mining both intra- and inter-image context.•We propose an Intra-image Multi-View Context (IMVC) network in MI3<math><msup is="true"><mrow is="true"></mrow><mrow is="true"><mn is="true">3</mn></mrow></msup></math>C, which contains the scene, surrounding, instance, and part branches to sufficiently extract intra-image context from multiple views and collaboratively integrate them for finer query-gallery matching.•We also propose an Inter-image Group Context Ranking (IGCR) algorithm in MI3<math><msup is="true"><mrow is="true"></mrow><mrow is="true"><mn is="true">3</mn></mrow></msup></math>C, which exploits group matching similarities as inter-image context to measure the holistic image matching similarity, yielding a more robust ranking among the whole gallery.•Extensive experiments on two popular person search benchmarks show that our method outperforms previous state-of-the-art methods by conspicuous margins. Specifically, for mAP and top-1 accuracy metrics, we achieve 96.7%/97.1% on the CUHK-SYSU dataset and 55.6%/90.8% on the PRW dataset.