In human detection algorithms, some recurrent issues are still challenging. Indeed, in a real application we have a lack of various human poses in our data. For example, in video surveillance it could be a significant issue to forget detecting a human in uncommon pose. We propose a compensation to these less frequent cases. We introduce a detection in rotated view to compensate limited various poses in our dataset. Instead of introducing a generative enhancement for every human body pose, which could be exhausting. Indeed, data annotation has become a difficulty for many applications. Moreover, a simple data augmentation is not efficient to generalize our training. We need a strategy to learn this new poses. So, we will introduce a process to compute pseudo labels in a sequence of rotated views. It avoids to use a clustering algorithm like dbscan [Chen et al.(2021)Chen, Lagadec, and Bremond.] or K-means to compute a pseudo label. A summary of theoretical results is described in a first section to illustrate our baseline. First of all, we will illustrate how we process pseudo label on various views by two mechanisms which have been called relaxation and scanning. We introduce a sequence of rotations of original view, to process a detection in a compensated view. In second step, we will learn this distortion created by the various views and construct an approach to distill original knowledge to a more generalized training.The main improvement of this approach is to propose an alternative solution at dbscan [Chen et al.(2021)Chen, Lagadec, and Bremond.] approach which is popular, but which is difficult to use in real applications. We obtain a significant result by improving accuracy in especially various pose datasets.

This work focuses on unsupervised representation learning in person re-identification (ReID). Recent self-supervised contrastive learning methods learn invariance by maximizing the representation similarity between two augmented views of a same image. However, traditional data augmentation may bring to the fore undesirable distortions on identity features, which is not always favorable in id-sensitive ReID tasks. In this article, we propose to replace traditional data augmentation with a generative adversarial network (GAN) that is targeted to generate augmented views for contrastive learning. A 3D mesh guided person image generator is proposed to disentangle a person image into id-related and id-unrelated features. Deviating from previous GAN-based ReID methods that only work in id-unrelated space (pose and camera style), we conduct GAN-based augmentation on both id-unrelated and id-related features. We further propose specific contrastive losses to help our network learn invariance from id-unrelated and id-related augmentations. By jointly training the generative and the contrastive modules, our method achieves new state-of-the-art unsupervised person ReID performance on mainstream large-scale benchmarks.

Recent self-supervised contrastive learning provides an effective approach for unsupervised person re-identification (ReID) by learning invariance from different views (transformed versions) of an input. In this paper, we incorporate a Generative Adversarial Network (GAN) and a contrastive learning module into one joint training framework. While the GAN provides online data augmentation for contrastive learning, the contrastive module learns view-invariant features for generation. In this context, we propose a mesh-based view generator. Specifically, mesh projections serve as references towards generating novel views of a person. In addition, we propose a view-invariant loss to facilitate contrastive learning between original and generated views. Deviating from previous GAN-based unsupervised ReID methods involving domain adaptation, we do not rely on a labeled source dataset, which makes our method more flexible. Extensive experimental results show that our method significantly outperforms state-of-the-art methods under both, fully unsupervised and unsupervised domain adaptive settings on several large scale ReID dat-sets. Source code and models are available under https://github.com/chenhao2345/GCL.

Unsupervised person re-identification (ReID) aims at learning discriminative identity features without annotations. Recently, self-supervised contrastive learning has gained increasing attention for its effectiveness in unsupervised representation learning. The main idea of instance contrastive learning is to match a same instance in different augmented views. However, the relationship between different instances has not been fully explored in previous contrastive methods, especially for instance-level contrastive loss. To address this issue, we propose Interinstance Contrastive Encoding (ICE) that leverages interinstance pairwise similarity scores to boost previous classlevel contrastive ReID methods. We first use pairwise similarity ranking as one-hot hard pseudo labels for hard instance contrast, which aims at reducing intra-class variance. Then, we use similarity scores as soft pseudo labels to enhance the consistency between augmented and original views, which makes our model more robust to augmentation perturbations. Experiments on several large-scale person ReID datasets validate the effectiveness of our proposed unsupervised method ICE, which is competitive with even supervised methods. Code is made available at https://github.com/chenhao2345/ICE.

The objective of unsupervised person re-identification (Re-ID) is to learn discriminative features without laborintensive identity annotations. State-of-the-art unsupervised Re-ID methods assign pseudo labels to unlabeled images in the target domain and learn from these noisy pseudo labels. Recently introduced Mean Teacher Model is a promising way to mitigate the label noise. However, during the training, self-ensembled teacher-student networks quickly converge to a consensus which leads to a local minimum. We explore the possibility of using an asymmetric structure inside neural network to address this problem. First, asymmetric branches are proposed to extract features in different manners, which enhances the feature diversity in appearance signatures. Then, our proposed cross-branch supervision allows one branch to get supervision from the other branch, which transfers distinct knowledge and enhances the weight diversity between teacher and student networks. Extensive experiments show that our proposed method can significantly surpass the performance of previous work on both unsupervised domain adaptation and fully unsupervised Re-ID tasks.

In Person Re-Identification (Re-ID) task, combining local and global features is a common strategy to overcome missing key parts and misalignment on models based only on global features. Using this combination, neural networks yield impressive performance in Re-ID task. Previous part-based models mainly focus on spatial partition strategies. Recently, operations on channel information, such as Group Normalization and Channel Attention, have brought significant progress to various visual tasks. However, channel partition has not drawn much attention in Person Re-ID. In this paper, we conduct a study to exploit the potential of channel partition in Re-ID task. Based on this study, we propose an end-to-end Spatial and Channel partition Representation network (SCR) in order to better exploit both spatial and channel information. Experiments conducted on three mainstream image-based evaluation protocols including Market-1501, DukeMTMC-ReID and CUHK03 and one video-based evaluation protocol MARS validate the performance of our model, which outperforms previous state-of- the-art in both single and cross domain Re-ID tasks.

In this paper we propose a modified version of the classical explicit snake (active contour) approach for the segmentation of irregular object shape in video images. In the classical approach the external evolution force is computed using the gradient of the image or the region information. In our original proposed method the external force is computed using the region information with two windows around the contour point. This new approach improves the speed of convergence and also the quality of segmentation especially for irregular objects shape(objects with sharp corners, deep concave/convex shape). As the contour points are attracted to the discontinuity of form, the modified snake is called “comer tracker snake”.