Zhuqing Jiang

Jiangbo Pei, Zhuqing Jiang, Aidong Men et al.

Source-free unsupervised domain adaptation (SFUDA) aims to learn a target domain model using unlabeled target data and the knowledge of a well-trained source domain model. Most previous SFUDA works focus on inferring semantics of target data based on the source knowledge. Without measuring the transferability of the source knowledge, these methods insufficiently exploit the source knowledge, and fail to identify the reliability of the inferred target semantics. However, existing transferability measurements require either source data or target labels, which are infeasible in SFUDA. To this end, firstly, we propose a novel Uncertainty-induced Transferability Representation (UTR), which leverages uncertainty as the tool to analyse the channel-wise transferability of the source encoder in the absence of the source data and target labels. The domain-level UTR unravels how transferable the encoder channels are to the target domain and the instance-level UTR characterizes the reliability of the inferred target semantics. Secondly, based on the UTR, we propose a novel Calibrated Adaption Framework (CAF) for SFUDA, including i)the source knowledge calibration module that guides the target model to learn the transferable source knowledge and discard the non-transferable one, and ii)the target semantics calibration module that calibrates the unreliable semantics. With the help of the calibrated source knowledge and the target semantics, the model adapts to the target domain safely and ultimately better. We verified the effectiveness of our method using experimental results and demonstrated that the proposed method achieves state-of-the-art performances on the three SFUDA benchmarks. Code is available at https://github.com/SPIresearch/UTR.

Yinsong Xu, Zhuqing Jiang, Aidong Men et al.

Existing domain adaptation methods assume that domain discrepancies are caused by a few discrete attributes and variations, e.g., art, real, painting, quickdraw, etc. We argue that this is not realistic as it is implausible to define the real-world datasets using a few discrete attributes. Therefore, we propose to investigate a new problem namely the Continuous Domain Adaptation (CDA) through the lens where infinite domains are formed by continuously varying attributes. Leveraging knowledge of two labeled source domains and several observed unlabeled target domains data, the objective of CDA is to learn a generalized model for whole data distribution with the continuous attribute. Besides the contributions of formulating a new problem, we also propose a novel approach as a strong CDA baseline. To be specific, firstly we propose a novel alternating training strategy to reduce discrepancies among multiple domains meanwhile generalize to unseen target domains. Secondly, we propose a continuity constraint when estimating the cross-domain divergence measurement. Finally, to decouple the discrepancy from the mini-batch size, we design a domain-specific queue to maintain the global view of the source domain that further boosts the adaptation performances. Our method is proven to achieve the state-of-the-art in CDA problem using extensive experiments. The code is available at https://github.com/SPIresearch/CDA.

Xiuqi Zheng, Yuhang Zhang, Haoran Zhang et al.

Adapting large pre-trained foundation models, e.g., SAM, for medical image segmentation remains a significant challenge. A crucial step involves the formulation of a series of specialized prompts that incorporate specific clinical instructions. Past works have been heavily reliant on a singular type of prompt for each instance, necessitating manual input of an ideally correct prompt, which is less efficient. To tackle this issue, we propose to utilize prompts of different granularity, which are sourced from original images to provide a broader scope of clinical insights. However, combining prompts of varying types can pose a challenge due to potential conflicts. In response, we have designed a coarse-to-fine mechanism, referred to as curriculum prompting, that progressively integrates prompts of different types. Through extensive experiments on three public medical datasets across various modalities, we demonstrate the effectiveness of our proposed approach, which not only automates the prompt generation process but also yields superior performance compared to other SAM-based medical image segmentation methods. Code is available at: https://github.com/AnnaZzz-zxq/Curriculum-Prompting.

Yupeng Jia, Haiyong Luo, Fang Zhao et al.

State estimation with sensors is essential for mobile robots. Due to different performance of sensors in different environments, how to fuse measurements of various sensors is a problem. In this paper, we propose a tightly coupled multi-sensor fusion framework, Lvio-Fusion, which fuses stereo camera, Lidar, IMU, and GPS based on the graph optimization. Especially for urban traffic scenes, we introduce a segmented global pose graph optimization with GPS and loop-closure, which can eliminate accumulated drifts. Additionally, we creatively use a actor-critic method in reinforcement learning to adaptively adjust sensors' weight. After training, actor-critic agent can provide the system better and dynamic sensors' weight. We evaluate the performance of our system on public datasets and compare it with other state-of-the-art methods, which shows that the proposed method achieves high estimation accuracy and robustness to various environments. And our implementations are open source and highly scalable.

Yiyun Zhao, Zhuqing Jiang, Aidong Men et al.

While deep Convolutional Neural Networks (CNNs) have shown extraordinary capability of modelling specific noise and denoising, they still perform poorly on real-world noisy images. The main reason is that the real-world noise is more sophisticated and diverse. To tackle the issue of blind denoising, in this paper, we propose a novel pyramid real image denoising network (PRIDNet), which contains three stages. First, the noise estimation stage uses channel attention mechanism to recalibrate the channel importance of input noise. Second, at the multi-scale denoising stage, pyramid pooling is utilized to extract multi-scale features. Third, the stage of feature fusion adopts a kernel selecting operation to adaptively fuse multi-scale features. Experiments on two datasets of real noisy photographs demonstrate that our approach can achieve competitive performance in comparison with state-of-the-art denoisers in terms of both quantitative measure and visual perception quality. Code is available at https://github.com/491506870/PRIDNet.

Tian He, Chen Yang, Jiawei Zhang et al.

Generative recommendation systems are increasingly adopted in local service platforms, where semantic relevance alone is insufficient without strict geographic feasibility. A key technical challenge lies in semantic ID (SID) tokenization, which directly impacts the recommendation performance. However, existing semantic codebooks neglect geographic constraints, often resulting in recommendations that are semantically relevant yet geographically unreachable. To address this limitation, we propose Pro-GEO, a Proximity-aware GEO-codebook. Pro-GEO establishes a geo-centroid local coordinate system to capture intra-cluster spatial relationships and a geo-rotary position encoding mechanism that models geographic proximity as orthogonal rotational transformations in the high-dimensional embedding. This design enables semantic and spatial signals to be jointly modeled in a balanced manner, without reducing geographic information to a weak auxiliary feature. Extensive experiments conducted on a large-scale industrial dataset reveal that Pro-GEO significantly outperforms state-of-the-art methods. In particular, Pro-GEO reduces the average geographic clustering distance by 45.60% and achieves a 1.87% improvement in Hit@50, highlighting its effectiveness for real-world local service recommendation.

Jiangbo Pei, Zhuqing Jiang, Aidong Men et al.

Single-camera-training person re-identification (SCT re-ID) aims to train a re-ID model using SCT datasets where each person appears in only one camera. The main challenge of SCT re-ID is to learn camera-invariant feature representations without cross-camera same-person (CCSP) data as supervision. Previous methods address it by assuming that the most similar person should be found in another camera. However, this assumption is not guaranteed to be correct. In this paper, we propose a Camera-Invariant Meta-Learning Network (CIMN) for SCT re-ID. CIMN assumes that the camera-invariant feature representations should be robust to camera changes. To this end, we split the training data into meta-train set and meta-test set based on camera IDs and perform a cross-camera simulation via meta-learning strategy, aiming to enforce the representations learned from the meta-train set to be robust to the meta-test set. With the cross-camera simulation, CIMN can learn camera-invariant and identity-discriminative representations even there are no CCSP data. However, this simulation also causes the separation of the meta-train set and the meta-test set, which ignores some beneficial relations between them. Thus, we introduce three losses: meta triplet loss, meta classification loss, and meta camera alignment loss, to leverage the ignored relations. The experiment results demonstrate that our method achieves comparable performance with and without CCSP data, and outperforms the state-of-the-art methods on SCT re-ID benchmarks. In addition, it is also effective in improving the domain generalization ability of the model.

Jianning Wu, Zhuqing Jiang, Shiping Wen et al.

For multimodal tasks, a good feature extraction network should extract information as much as possible and ensure that the extracted feature embedding and other modal feature embedding have an excellent mutual understanding. The latter is often more critical in feature fusion than the former. Therefore, selecting the optimal feature extraction network collocation is a very important subproblem in multimodal tasks. Most of the existing studies ignore this problem or adopt an ergodic approach. This problem is modeled as an optimization problem in this paper. A novel method is proposed to convert the optimization problem into an issue of comparative upper bounds by referring to the general practice of extreme value conversion in mathematics. Compared with the traditional method, it reduces the time cost. Meanwhile, aiming at the common problem that the feature similarity and the feature semantic similarity are not aligned in the multimodal time-series problem, we refer to the idea of contrast learning and propose a multimodal time-series contrastive loss(MTSC). Based on the above issues, We demonstrated the feasibility of our approach in the audio-visual video parsing task. Substantial analyses verify that our methods promote the fusion of different modal features.

Ting Pan, Zhuqing Jiang, Jianan Han et al.

Video prediction is a challenging task with wide application prospects in meteorology and robot systems. Existing works fail to trade off short-term and long-term prediction performances and extract robust latent dynamics laws in video frames. We propose a two-branch seq-to-seq deep model to disentangle the Taylor feature and the residual feature in video frames by a novel recurrent prediction module (TaylorCell) and residual module. TaylorCell can expand the video frames' high-dimensional features into the finite Taylor series to describe the latent laws. In TaylorCell, we propose the Taylor prediction unit (TPU) and the memory correction unit (MCU). TPU employs the first input frame's derivative information to predict the future frames, avoiding error accumulation. MCU distills all past frames' information to correct the predicted Taylor feature from TPU. Correspondingly, the residual module extracts the residual feature complementary to the Taylor feature. On three generalist datasets (Moving MNIST, TaxiBJ, Human 3.6), our model outperforms or reaches state-of-the-art models, and ablation experiments demonstrate the effectiveness of our model in long-term prediction.

Zhuqing Jiang, Chang Liu, Ya'nan Wang et al.

With the goal of tuning up the brightness, low-light image enhancement enjoys numerous applications, such as surveillance, remote sensing and computational photography. Images captured under low-light conditions often suffer from poor visibility and blur. Solely brightening the dark regions will inevitably amplify the blur, thus may lead to detail loss. In this paper, we propose a simple yet effective two-stream framework named NEID to tune up the brightness and enhance the details simultaneously without introducing many computational costs. Precisely, the proposed method consists of three parts: Light Enhancement (LE), Detail Refinement (DR) and Feature Fusing (FF) module, which can aggregate composite features oriented to multiple tasks based on channel attention mechanism. Extensive experiments conducted on several benchmark datasets demonstrate the efficacy of our method and its superiority over state-of-the-art methods.

Ya'nan Wang, Zhuqing Jiang, Chang Liu et al.

It is suggested that low-light image enhancement realizes one-to-many mapping since we have different definitions of NORMAL-light given application scenarios or users' aesthetic. However, most existing methods ignore subjectivity of the task, and simply produce one result with fixed brightness. This paper proposes a neural network for multi-level low-light image enhancement, which is user-friendly to meet various requirements by selecting different images as brightness reference. Inspired by style transfer, our method decomposes an image into two low-coupling feature components in the latent space, which allows the concatenation feasibility of the content components from low-light images and the luminance components from reference images. In such a way, the network learns to extract scene-invariant and brightness-specific information from a set of image pairs instead of learning brightness differences. Moreover, information except for the brightness is preserved to the greatest extent to alleviate color distortion. Extensive results show strong capacity and superiority of our network against existing methods.

Zhuqing Jiang, Haotian Li, Liangjie Liu et al.

Self-regularized low-light image enhancement does not require any normal-light image in training, thereby freeing from the chains on paired or unpaired low-/normal-images. However, existing methods suffer color deviation and fail to generalize to various lighting conditions. This paper presents a novel self-regularized method based on Retinex, which, inspired by HSV, preserves all colors (Hue, Saturation) and only integrates Retinex theory into brightness (Value). We build a reflectance estimation network by restricting the consistency of reflectances embedded in both the original and a novel random disturbed form of the brightness of the same scene. The generated reflectance, which is assumed to be irrelevant of illumination by Retinex, is treated as enhanced brightness. Our method is efficient as a low-light image is decoupled into two subspaces, color and brightness, for better preservation and enhancement. Extensive experiments demonstrate that our method outperforms multiple state-of-the-art algorithms qualitatively and quantitatively and adapts to more lighting conditions.

Fan Zhou, Zhuqing Jiang, Changjian Shui et al.

Generalizing knowledge to unseen domains, where data and labels are unavailable, is crucial for machine learning models. We tackle the domain generalization problem to learn from multiple source domains and generalize to a target domain with unknown statistics. The crucial idea is to extract the underlying invariant features across all the domains. Previous domain generalization approaches mainly focused on learning invariant features and stacking the learned features from each source domain to generalize to a new target domain while ignoring the label information, which will lead to indistinguishable features with an ambiguous classification boundary. For this, one possible solution is to constrain the label-similarity when extracting the invariant features and to take advantage of the label similarities for class-specific cohesion and separation of features across domains. Therefore we adopt optimal transport with Wasserstein distance, which could constrain the class label similarity, for adversarial training and also further deploy a metric learning objective to leverage the label information for achieving distinguishable classification boundary. Empirical results show that our proposed method could outperform most of the baselines. Furthermore, ablation studies also demonstrate the effectiveness of each component of our method.

Qiulin Zhang, Zhuqing Jiang, Qishuo Lu et al.

Many effective solutions have been proposed to reduce the redundancy of models for inference acceleration. Nevertheless, common approaches mostly focus on eliminating less important filters or constructing efficient operations, while ignoring the pattern redundancy in feature maps. We reveal that many feature maps within a layer share similar but not identical patterns. However, it is difficult to identify if features with similar patterns are redundant or contain essential details. Therefore, instead of directly removing uncertain redundant features, we propose a \textbf{sp}lit based \textbf{conv}olutional operation, namely SPConv, to tolerate features with similar patterns but require less computation. Specifically, we split input feature maps into the representative part and the uncertain redundant part, where intrinsic information is extracted from the representative part through relatively heavy computation while tiny hidden details in the uncertain redundant part are processed with some light-weight operation. To recalibrate and fuse these two groups of processed features, we propose a parameters-free feature fusion module. Moreover, our SPConv is formulated to replace the vanilla convolution in a plug-and-play way. Without any bells and whistles, experimental results on benchmarks demonstrate SPConv-equipped networks consistently outperform state-of-the-art baselines in both accuracy and inference time on GPU, with FLOPs and parameters dropped sharply.

Yue Lu, Yun Zhou, Zhuqing Jiang et al.

Convolutional neural networks (CNNs) have demonstrated superior performance in super-resolution (SR). However, most CNN-based SR methods neglect the different importance among feature channels or fail to take full advantage of the hierarchical features. To address these issues, this paper presents a novel recursive unit. Firstly, at the beginning of each unit, we adopt a compact channel attention mechanism to adaptively recalibrate the channel importance of input features. Then, the multi-level features, rather than only deep-level features, are extracted and fused. Additionally, we find that it will force our model to learn more details by using the learnable upsampling method (i.e., transposed convolution) only on residual branch (instead of using it both on residual branch and identity branch) while using the bicubic interpolation on the other branch. Analytic experiments show that our method achieves competitive results compared with the state-of-the-art methods and maintains faster speed as well.