Unsupervised person re-identification (re-ID) attracts increasing attention due to its practical applications in industry. State-of-the-art unsupervised re-ID methods train the neural networks using a memory-based non-parametric softmax loss. They store the pre-computed instance feature vectors inside the memory, assign pseudo labels to them using clustering algorithm, and compare the query instances to the cluster using a form of contrastive loss. During training, the instance feature vectors are updated. However, due to the varying cluster size, the updating progress for each cluster is inconsistent. To solve this problem, we present Cluster Contrast which stores feature vectors and computes contrast loss in the cluster level. We demonstrate that the inconsistency problem for cluster feature representation can be solved by the cluster-level memory dictionary. By straightforwardly applying Cluster Contrast to a standard unsupervised re-ID pipeline, it achieves considerable improvements of 9.5%, 7.5%, 6.6% compared to state-ofthe-art purely unsupervised re-ID methods and 5.1%, 4.0%, 6.5% mAP compared to the state-of-the-art unsupervised domain adaptation re-ID methods on the Market, Duke, and MSMT17 datasets.

No abstract

No abstract

The fusion of the heterogeneous sensors can greatly improve the environmental perception ability of mobile robots. And that the primary difficulty of heterogeneous sensors fusion is the calibration of depth scan information and plane image information for a laser rangefinder and a camera. Firstly, a coordinate transformation method from a laser rangefinder coordinates system to an optical image plane is given, and then the calibration of the camera's intrinsic parameters is achieved by “Camera Calibration Toolbox‘. Secondly, the intrinsic and extrinsic parameters are separated for calibration are proposed and compared, in which the characteristic parameters' identification is according to some characteristic points on the intersection line. Then Gaussian elimination is utilized for the initial value. Furthermore, the parameters' optimization using the non-linear least square and non-linear Gauss-Newton methods is devised for different constraints. Finally, the simulated and real experimental results demonstrate the reliability and effectiveness of extrinsic and intrinsic parameters' separated calibration, meanwhile, the real-time analysis is achieved for robotic multi-sensor fusion.

No abstract