Qin Xu

Jiaming Liu, Chi-Hao Wu, Yuzhi Wang et al.

In this paper, we present new data pre-processing and augmentation techniques for DNN-based raw image denoising. Compared with traditional RGB image denoising, performing this task on direct camera sensor readings presents new challenges such as how to effectively handle various Bayer patterns from different data sources, and subsequently how to perform valid data augmentation with raw images. To address the first problem, we propose a Bayer pattern unification (BayerUnify) method to unify different Bayer patterns. This allows us to fully utilize a heterogeneous dataset to train a single denoising model instead of training one model for each pattern. Furthermore, while it is essential to augment the dataset to improve model generalization and performance, we discovered that it is error-prone to modify raw images by adapting augmentation methods designed for RGB images. Towards this end, we present a Bayer preserving augmentation (BayerAug) method as an effective approach for raw image augmentation. Combining these data processing technqiues with a modified U-Net, our method achieves a PSNR of 52.11 and a SSIM of 0.9969 in NTIRE 2019 Real Image Denoising Challenge, demonstrating the state-of-the-art performance. Our code is available at https://github.com/Jiaming-Liu/BayerUnifyAug.

Jiahui Wang, Qin Xu, Bo Jiang et al.

Pre-trained large vision-language models (VLMs) like CLIP demonstrate impressive generalization ability. Existing prompt-based and adapter-based works have made significant progress in fine-tuning VLMs but still face the challenges of maintaining strong generalization abilities, particularly towards unseen new classes. This limitation partly arises from these methods treating all tokens of the image and text encoder equally, which can lead to overfitting on less informative features (e.g., background noise, template words) and degrade the general representations that are crucial for novel concept recognition. To address this issue, we propose Dynamic Rank Adaptation (DRA), a novel adapter variant method, designed specifically to enhance new class generalization. DRA dynamically allocates adaptation ranks based on the importance of features during training to preserve general knowledge. DRA first employs token importance grouping, using sequence attention to evaluate and group tokens by their importance. Then, we adopt rank adaptation according to the importance of each token group dynamically by assigning higher feature ranks to the more important tokens. Also, we design a new channel response mechanism to prioritize the preservation and adaptation of feature channels identified as the most informative for each instance. In addition, a L1 regularization term is introduced to stabilize the training. Extensive experiments demonstrate the effectiveness and superiority of our proposed DRA over existing works, especially on enhancing the performance of new classes on various benchmarks, including base-new classes, cross-datasets evaluation and domain generalization. The source code will be published after the paper is received.

Yuhe Ding, Bo Jiang, Aihua Zheng et al.

Vision language models (VLMs) like CLIP show stellar zero-shot capability on classification benchmarks. However, selecting the VLM with the highest performance on the unlabeled downstream task is non-trivial. Existing VLM selection methods focus on the class-name-only setting, relying on a supervised large-scale dataset and large language models, which may not be accessible or feasible during deployment. This paper introduces the problem of \textbf{unsupervised vision-language model selection}, where only unsupervised downstream datasets are available, with no additional information provided. To solve this problem, we propose a method termed Visual-tExtual Graph Alignment (VEGA), to select VLMs without any annotations by measuring the alignment of the VLM between the two modalities on the downstream task. VEGA is motivated by the pretraining paradigm of VLMs, which aligns features with the same semantics from the visual and textual modalities, thereby mapping both modalities into a shared representation space. Specifically, we first construct two graphs on the vision and textual features, respectively. VEGA is then defined as the overall similarity between the visual and textual graphs at both node and edge levels. Extensive experiments across three different benchmarks, covering a variety of application scenarios and downstream datasets, demonstrate that VEGA consistently provides reliable and accurate estimates of VLMs' performance on unlabeled downstream tasks.

Yuzhi Wang, Haibin Huang, Qin Xu et al.

Deep learning-based image denoising approaches have been extensively studied in recent years, prevailing in many public benchmark datasets. However, the stat-of-the-art networks are computationally too expensive to be directly applied on mobile devices. In this work, we propose a light-weight, efficient neural network-based raw image denoiser that runs smoothly on mainstream mobile devices, and produces high quality denoising results. Our key insights are twofold: (1) by measuring and estimating sensor noise level, a smaller network trained on synthetic sensor-specific data can out-perform larger ones trained on general data; (2) the large noise level variation under different ISO settings can be removed by a novel k-Sigma Transform, allowing a small network to efficiently handle a wide range of noise levels. We conduct extensive experiments to demonstrate the efficiency and accuracy of our approach. Our proposed mobile-friendly denoising model runs at ~70 milliseconds per megapixel on Qualcomm Snapdragon 855 chipset, and it is the basis of the night shot feature of several flagship smartphones released in 2019.

Xu Qin, Zhilin Wang

Images captured under complicated rain conditions often suffer from noticeable degradation of visibility. The rain models generally introduce diversity visibility degradation, which includes rain streak, rain drop as well as rain mist. Numerous existing single image deraining methods focus on the only one type rain model, which does not have strong generalization ability. In this paper, we propose a novel end-to-end Neuron Attention Stage-by-Stage Net (NASNet), which can solve all types of rain model tasks efficiently. For one thing, we pay more attention on the Neuron relationship and propose a lightweight Neuron Attention (NA) architectural mechanism. It can adaptively recalibrate neuron-wise feature responses by modelling interdependencies and mutual influence between neurons. Our NA architecture consists of Depthwise Conv and Pointwise Conv, which has slight computation cost and higher performance than SE block by our contrasted experiments. For another, we propose a stage-by-stage unified pattern network architecture, the stage-by-stage strategy guides the later stage by incorporating the useful information in previous stage. We concatenate and fuse stage-level information dynamically by NA module. Extensive experiments demonstrate that our proposed NASNet significantly outperforms the state-of-theart methods by a large margin in terms of both quantitative and qualitative measures on all six public large-scale datasets for three rain model tasks.

Xu Qin, Zhilin Wang, Yuanchao Bai et al.

In this paper, we propose an end-to-end feature fusion at-tention network (FFA-Net) to directly restore the haze-free image. The FFA-Net architecture consists of three key components: 1) A novel Feature Attention (FA) module combines Channel Attention with Pixel Attention mechanism, considering that different channel-wise features contain totally different weighted information and haze distribution is uneven on the different image pixels. FA treats different features and pixels unequally, which provides additional flexibility in dealing with different types of information, expanding the representational ability of CNNs. 2) A basic block structure consists of Local Residual Learning and Feature Attention, Local Residual Learning allowing the less important information such as thin haze region or low-frequency to be bypassed through multiple local residual connections, let main network architecture focus on more effective information. 3) An Attention-based different levels Feature Fusion (FFA) structure, the feature weights are adaptively learned from the Feature Attention (FA) module, giving more weight to important features. This structure can also retain the information of shallow layers and pass it into deep layers. The experimental results demonstrate that our proposed FFA-Net surpasses previous state-of-the-art single image dehazing methods by a very large margin both quantitatively and qualitatively, boosting the best published PSNR metric from 30.23db to 36.39db on the SOTS indoor test dataset. Code has been made available at GitHub.

Tieyuan Lu, Xinlin Zhang, Yihui Huang et al.

Magnetic resonance imaging has been widely applied in clinical diagnosis, however, is limited by its long data acquisition time. Although imaging can be accelerated by sparse sampling and parallel imaging, achieving promising reconstruction images with a fast reconstruction speed remains a challenge. Recently, deep learning approaches have attracted a lot of attention for its encouraging reconstruction results but without a proper interpretability. In this letter, to enable high-quality image reconstruction for the parallel magnetic resonance imaging, we design the network structure from the perspective of sparse iterative reconstruction and enhance it with the residual structure. The experimental results of a public knee dataset show that compared with the optimization-based method and the latest deep learning parallel imaging methods, the proposed network has less error in reconstruction and is more stable under different acceleration factors.

Xinlin Zhang, Hengfa Lu, Di Guo et al.

The boom of non-uniform sampling and compressed sensing techniques dramatically alleviates the lengthy data acquisition problem of magnetic resonance imaging. Sparse reconstruction, thanks to its fast computation and promising performance, has attracted researchers to put numerous efforts on it and has been adopted in commercial scanners. To perform sparse reconstruction, choosing a proper algorithm is essential in providing satisfying results and saving time in tuning parameters. The pFISTA, a simple and efficient algorithm for sparse reconstruction, has been successfully extended to parallel imaging. However, its convergence criterion is still an open question. And the existing convergence criterion of single-coil pFISTA cannot be applied to the parallel imaging pFISTA, which, therefore, imposes confusions and difficulties on users about determining the only parameter - step size. In this work, we provide the guaranteed convergence analysis of the parallel imaging version pFISTA to solve the two well-known parallel imaging reconstruction models, SENSE and SPIRiT. Along with the convergence analysis, we provide recommended step size values for SENSE and SPIRiT reconstructions to obtain fast and promising reconstructions. Experiments on in vivo brain images demonstrate the validity of the convergence criterion. Besides, experimental results show that compared to using backtracking and power iteration to determine the step size, our recommended step size achieves more than five times acceleration in reconstruction time in most tested cases.

Yifan Sun, Qin Xu, Yali Li et al.

This paper considers a realistic problem in person re-identification (re-ID) task, i.e., partial re-ID. Under partial re-ID scenario, the images may contain a partial observation of a pedestrian. If we directly compare a partial pedestrian image with a holistic one, the extreme spatial misalignment significantly compromises the discriminative ability of the learned representation. We propose a Visibility-aware Part Model (VPM), which learns to perceive the visibility of regions through self-supervision. The visibility awareness allows VPM to extract region-level features and compare two images with focus on their shared regions (which are visible on both images). VPM gains two-fold benefit toward higher accuracy for partial re-ID. On the one hand, compared with learning a global feature, VPM learns region-level features and benefits from fine-grained information. On the other hand, with visibility awareness, VPM is capable to estimate the shared regions between two images and thus suppresses the spatial misalignment. Experimental results confirm that our method significantly improves the learned representation and the achieved accuracy is on par with the state of the art.