We introduce AiT, a unified output representation for various vision tasks, which is a crucial step towards general-purpose vision task solvers. Despite the challenges posed by the high-dimensional and task-specific outputs, we showcase the potential of using discrete representation (VQVAE) to model the dense outputs of many computer vision tasks as a sequence of discrete tokens. This is inspired by the established ability of VQ-VAE to conserve the structures spanning multiple pixels using few discrete codes. To that end, we present a modified shallower architecture for VQ-VAE that improves efficiency while keeping prediction accuracy. Our approach also incorporates uncertainty into the decoding process by using a soft fusion of the codebook entries, providing a more stable training process, which notably improved prediction accuracy. Our evaluation of AiT on depth estimation and instance segmentation tasks, with both continuous and discrete labels, demonstrates its superiority compared to other unified models. The code and models are available at https://github.com/SwinTransformer/AiT.

Active learning is a label-efficient machine learning method that actively selects the most valuable unlabeled samples to annotate. Active learning focuses on achieving the best possible performance while using as few, high-quality sample annotations as possible. Recently, active learning achieved promotion combined with deep learning-based methods, which are named deep active learning methods in this paper. Deep active learning plays a crucial role in computer vision tasks, especially in label-insensitive scenarios, such as hard-to-label tasks (medical images analysis) and time-consuming tasks (autonomous driving). However, deep active learning still has some challenges, such as unstable performance and dirty data, which are future research trends. Compared with other reviews on deep active learning, our work introduced the deep active learning from computer vision-related methodologies and corresponding applications. The expected audience of this vision-friendly survey are researchers who are working in computer vision but willing to utilize deep active learning methods to solve vision problems. Specifically, this review systematically focuses on the details of methods, applications, and challenges in vision tasks, and we also introduce the classic theories, strategies, and scenarios of active learning in brief.

Mainstream unsupervised domain adaptation (UDA) methods align feature distributions across different domains via adversarial learning. However, most of them focus on global distribution alignment, ignoring the fine-grained domain discrepancy. Besides, they generally require auxiliary models, bringing extra computation costs. To tackle these issues, this study proposes an UDA method that differentiates individual samples without the help of extra models. To this end, we introduce a novel discrepancy metric, termed style discrepancy, to distinguish different target samples. We also propose a paradigm for adversarial style discrepancy minimization (ASDM). Specifically, we fix the parameters of the feature extractor and maximize style discrepancy to update the classifier, which helps detect more hard samples. Adversely, we fix the parameters of the classifier and minimize the style discrepancy to update the feature extractor, pushing those hard samples near the support of the source distribution. Such adversary helps to progressively detect and adapt more hard samples, leading to fine-grained domain adaptation. Experiments on different UDA tasks validate the effectiveness of ASDM. Overall, without any extra models, ASDM reaches a 46.9% mIoU in the GTA5 to Cityscapes benchmark and an 84.7% accuracy in the VisDA-2017 benchmark, outperforming many existing adversarial-learning-based methods.

Self-supervised pre-training has become the priory choice to establish reliable neural networks for automated recognition of massive biomedical microscopy images, which are routinely annotation-free, without semantics, and without guarantee of quality. Note that this paradigm is still at its infancy and limited by closely related open issues: (1) how to learn robust representations in an unsupervised manner from unlabeled biomedical microscopy images of low diversity in samples? and (2) how to obtain the most significant representations demanded by a high-quality segmentation? Aiming at these issues, this study proposes a knowledge-based learning framework (TOWER) towards enhanced recognition of biomedical microscopy images, which works in three phases by synergizing contrastive learning and generative learning methods: (1) Sample Space Diversification: Reconstructive proxy tasks have been enabled to embed a priori knowledge with context highlighted to diversify the expanded sample space; (2) Enhanced Representation Learning: Informative noise-contrastive estimation loss regularizes the encoder to enhance representation learning of annotation-free images; (3) Correlated Optimization: Optimization operations in pre-training the encoder and the decoder have been correlated via image restoration from proxy tasks, targeting the need for semantic segmentation. Experiments have been conducted on public datasets of biomedical microscopy images against the state-of-the-art counterparts (e.g., SimCLR and BYOL), and results demonstrate that: TOWER statistically excels in all self-supervised methods, achieving a Dice improvement of 1.38 percentage points over SimCLR. TOWER also has potential in multi-modality medical image analysis and enables label-efficient semi-supervised learning, e.g., reducing the annotation cost by up to 99% in pathological classification.

Malignant lesions are a huge threat to human health and have a high mortality rate. Locating the contour of organs is a preparation step, and it helps doctors diagnose correctly. Therefore, there is an urgent clinical need for a segmentation model specifically designed for medical imaging. However, most current medical image segmentation models directly migrate from natural image segmentation models, thus ignoring some characteristic features for medical images, such as false positive phenomena and the blurred boundary problem in 3D volume data. The research on organ segmentation models for medical images is still challenging and demanding. As a consequence, we redesign a 3D convolutional neural network (CNN) based on 3D U-Net and adopted the render method from computer graphics for 3D medical images segmentation, named Render 3D U-Net. This network adapts a subdivision-based point-sampling method to replace the original upsampling method for rendering high-quality boundaries. Besides, Render 3D U-Net integrates the point-sampling method into 3D ANU-Net architecture under deep supervision. Meanwhile, to reduce false positive phenomena in clinical diagnosis and to achieve more accurate segmentation, Render 3D U-Net specially designs a module for screening false positive. Finally, three public challenge datasets (MICCAI 2017 LiTS, MICCAI 2019 KiTS, and ISBI 2019 segTHOR) were selected as experiment datasets and to evaluate the performance on target organs. Compared with other models, Render 3D U-Net improved the performance on both overall organ and boundary in the CT image segmentation tasks, including in the liver, kidney, and heart.