This paper addresses the re-ranking problem in person re-identification (re-ID), aiming to improve the accuracy of re-ID results. The authors propose a k-reciprocal encoding method to re-rank the initial ranking list. The hypothesis is that if a gallery image is similar to the probe in the k-reciprocal nearest neighbors, it is more likely to be a true match. The method encodes the k-reciprocal nearest neighbors into a single vector, which is used to compute the Jaccard distance between images. The final distance is a combination of the original distance and the Jaccard distance. The proposed method does not require human interaction or labeled data and can be applied to large-scale datasets. Experiments on Market-1501, CUHK03, MARS, and PRW datasets demonstrate the effectiveness of the method, achieving state-of-the-art accuracy on Market-1501 in both rank-1 and mAP. The contributions of the paper include a k-reciprocal feature, an automatic and unsupervised re-ranking method, and significant improvements in re-ID performance on multiple datasets.This paper addresses the re-ranking problem in person re-identification (re-ID), aiming to improve the accuracy of re-ID results. The authors propose a k-reciprocal encoding method to re-rank the initial ranking list. The hypothesis is that if a gallery image is similar to the probe in the k-reciprocal nearest neighbors, it is more likely to be a true match. The method encodes the k-reciprocal nearest neighbors into a single vector, which is used to compute the Jaccard distance between images. The final distance is a combination of the original distance and the Jaccard distance. The proposed method does not require human interaction or labeled data and can be applied to large-scale datasets. Experiments on Market-1501, CUHK03, MARS, and PRW datasets demonstrate the effectiveness of the method, achieving state-of-the-art accuracy on Market-1501 in both rank-1 and mAP. The contributions of the paper include a k-reciprocal feature, an automatic and unsupervised re-ranking method, and significant improvements in re-ID performance on multiple datasets.