YXYuxuan Xiao

Yuxuan Xiao

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Yuxuan Xiao has a total of 5 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Siamese Feature Decoupling and Adaptive Prototype Alignment for Clothes Changed Person Re-Identification

    by Yuxuan Xiao, Jian Yang, Shanshan Zhang

    The core to tackle clothes changed person re-identification is to extract clothes-irrelevant features. Previous methods typically decoupled the entire image features into clothes-relevant and ID-relevant ones via an auxiliary task of clothes classification, yet in this way, it is not guaranteed that the ID-relevant features fully get rid of clothes-relevant information, as the entire image feature also includes other information, such as posture. To enhance the purity of ID-relevant features, this letter proposes a simple yet effective siamese feature decoupling and adaptive prototype alignment framework (Decoupling and Alignment, D${\&}$A), which involves a reference branch and an alignment branch. Specifically, the reference branch first uses human parsing to remove clothes information at the image level to ensure that the ID prototype subsequently extracted by the image encoder is purely ID-relevant without clothes; the alignment branch excludes clothes information from the ID-relevant features by dual-token decoupling, and then aligns the ID-relevant features with the ID prototypes in the reference branch to enhance the purity of the identity information obtained after decoupling. Furthermore, we introduce an adaptive update strategy to update the ID prototypes to prevent the ID prototypes and the decoupled features from shifting in feature space. Extensive experiments on widely used benchmarks (PRCC, LTCC, and VC-Clothes) validate the effectiveness of our method, which achieves new state-of-the-art results across most metrics.


  • Domain Invariant Noise-Tolerant Learning for Unsupervised Cross-Domain Person Re-ID

    by Yuxuan Xiao, Jian Yang, Shanshan Zhang

    In this paper, we focus on tackling the task of unsupervised cross-domain adaptation on person re-identification (UDA re-ID). Most previous methods formulate UDA re-ID as a semi-supervised learning problem and aim to generate pseudo-labels on the target domain for training. However, the quality of pseudo-labels is far from satisfactory due to large domain gaps. To address this problem, in this work, we propose to improve the quality of pseudo-labels by handling domain shifts. Specifically, we narrow down the domain gap via domain style normalization and allow the model to behave more robustly to domain shifts by applying different perturbations on the target data. Furthermore, we point out that noises are inevitable in pseudo-labels and propose a noise-tolerant learning paradigm, which consists of positive learning on latent clean pseudo-labels and negative learning on latent noisy pseudo-labels. We conduct experiments under different adaptation scenarios, including syntheticreal and realreal, and demonstrate that our proposed Domain Invariant Noise-Tolerant Learning (DINTL) method achieves state-of-the-art results without additional parameters.


  • Semantic‐aware automatic image colorization via unpaired cycle‐consistent self‐supervised network

    by Yuxuan Xiao, Aiwen Jiang, Changhong Liu, Mingwen Wang

    Automatic image colorization without manual interventions is an ill-conditioned and inherently ambiguous problem. Most of existing methods focus on formulating colorization as a regression problem and learn parametric mappings from grayscale to color through deep neural networks. Due to the multimodalities of color-grayscale space, in many applications, it is not required to recover exact ground-truth color. Pair-wise pixel-to-pixel learning-based algorithms lack rationality. Techniques such as color space conversion techniques are then proposed to avoid such direct pixel learning. However, the coloring results after color space conversion are blunt and unnatural. In this paper, we hold viewpoints that a reasonable solution is to generate some colorized result that looks natural. No matter what color a region is to be assigned, the colorized region should be semantically and spatially consistent. In this paper, we propose an effective semantic-aware automatic colorization model via unpaired cycle-consistent self-supervised network. Low-level monochrome loss, perceptual identity loss and high-level semantic-consistence loss, together with adversarial loss, are introduced to guide network self-training. We train and test our model on randomly selected subsets from PASCAL VOC 2012. The experimental results including human subjective studies demonstrate that, compared with state-of-the-art methods, our proposed model can achieve more convincing and superior results. Relevant source code is available at https://github.com/YuSuen/ACCycleGAN.


  • Single Image Colorization Via Modified Cyclegan

    by Yuxuan Xiao, Aiwen Jiang, Changhong Liu, Mingwen Wang

    In this paper, we focus on automatically colorizing single grayscale image without manual interventions. Most of existing methods tried to accurately restore unknown ground-truth colors and require paired training data for model optimization. However, the ideal restoration objective and strict training constraints limited their performance. Inspired by CycleGAN, we formulate the process of colorization as image-to-image translation and propose an effective color-CycleGAN solution. High-level semantic identity loss and low-level color loss are additionally suggested for model optimization. Our method allows using unpaired images for training and direct prediction in rgb color space, which makes training data collection much easier and more general. We train our model on randomly selected PASCAL VOC 2007 images. All ablation study on loss function and comparisons with state-of-the-art methods are performed on grayscale SUN data. The experiment results show that our improvements on training loss could achieve better content consistence and generate better reasonable colors with less artifacts. Moreover, due to the bidirectional nature of our model, our proposed method provides a by-product that gives an excellent alternative way on color image graying.