Domain generalized medical image segmentation requires models to learn from multiple source domains and generalize well to arbitrary unseen target domain. Such a task is both technically challenging and clinically practical, due to the domain shift problem (i.e., images are collected from different hospitals and scanners). Existing methods focused on either learning shape-invariant representation or reaching consensus among the source domains. An ideal generalized representation is supposed to show similar pattern responses within the same channel for cross-domain images. However, to deal with the significant distribution discrepancy, the network tends to capture similar patterns by multiple channels, while different cross-domain patterns are also allowed to rest in the same channel. To address this issue, we propose to leverage channel-wise decoupled deep features as queries. With the aid of cross-attention mechanism, the long-range dependency between deep and shallow features can be fully mined via self-attention and then guides the learning of generalized representation. Besides, a relaxed deep whitening transformation is proposed to learn channel-wise decoupled features in a feasible way. The proposed decoupled fea- ture query (DFQ) scheme can be seamlessly integrate into the Transformer segmentation model in an end-to-end manner. Extensive experiments show its state-of-the-art performance, notably outperforming the runner-up by 1.31% and 1.98% with DSC metric on generalized fundus and prostate benchmarks, respectively. Source code is available at https://github.com/BiQiWHU/DFQ.

The rapid development of Vision Foundation Model (VFM) brings inherent out-domain generalization for a variety of down-stream tasks. Among them, domain generalized semantic segmentation (DGSS) holds unique challenges as the cross-domain images share common pixel-wise content information but vary greatly in terms of the style. In this paper, we present a novel Spectral-dEcomposed Token (SET) learning framework to advance the frontier. Delving into further than existing fine-tuning token & frozen backbone paradigm, the proposed SET especially focuses on the way learning style-invariant features from these learnable tokens. Particularly, the frozen VFM features are first decomposed into the phase and amplitude components in the frequency space, which mainly contain the information of content and style, respectively, and then separately processed by learnable tokens for task-specific information extraction. Particularly, the frozen VFM features are first decomposed into the phase and amplitude components in the frequency space, which mainly contain the information of content and style, respectively, and then separately processed by learnable tokens for task-specific information extraction.After the decomposition, style variation primarily impacts the token-based feature enhancement within the amplitude branch. To address this issue, we further develop an attention optimization method to bridge the gap between style-affected representation and static tokens during inference. Extensive cross-domain experiments show its state-of-the-art performance.

Generic image classification has been widely studied in the past decade. However, for the bird-view aerial images, aerial scene classification remains challenging due to the dramatic variation of the scale and object size. Existing methods usually learn the aerial scene representation from the convolutional neural networks (CNN), which focus on the local response of an image. In contrast, the recently-developed vision transformers (ViT) can learn stronger global representation for aerial scenes, but are not qualified enough to highlight the key objects in an aerial scene due to the dramatic size and scale variation. To address this challenge, in this paper, we propose a local-global interactive vision transformer (LG-ViT) for this task. It is based on our deliberately designed local-global feature interactive learning scheme, which intends to jointly utilize the local-wise and global-wise feature representations. To realize the learning scheme in an end-to-end manner, the proposed LG-ViT consists of three key components, namely, local-global feature extraction, local-global feature interaction, and local-global semantic constraints. Extensive experiments on three aerial scene classification benchmarks, namely, UCM, AID and NWPU, demonstrate the effectiveness of the proposed LG-ViT against the state-of-the-art methods. The effectiveness of each component and generalization capability is also validated.

High-quality annotation of fine-grained visual categories demands great expert knowledge, which is taxing and time consuming. Alternatively, learning fine-grained visual representation from enormous unlabeled images (e.g., species, brands) by self-supervised learning becomes a feasible solution. However, recent investigations find that existing self-supervised learning methods are less qualified to represent fine-grained categories. The bottleneck lies in that the pre-trained class-agnostic representation is built from every patch-wise embedding, while fine-grained categories are only determined by several key patches of an image. In this paper, we propose a Cross-level Multi-instance Distillation (CMD) framework to tackle this challenge. Our key idea is to consider the importance of each image patch in determining the fine-grained representation by multiple instance learning. To comprehensively learn the relation between informative patches and fine-grained semantics, the multi-instance knowledge distillation is implemented on both the region/image crop pairs from the teacher and student net, and the region-image crops inside the teacher / student net, which we term as intra-level multi-instance distillation and inter-level multi-instance distillation. Extensive experiments on several commonly used datasets, including CUB-200-2011, Stanford Cars and FGVC Aircraft, demonstrate that the proposed method outperforms the contemporary methods by up to 10.14% and existing state-of-the-art self-supervised learning approaches by up to 19.78% on both top-1 accuracy and Rank-1 retrieval metric. Source code is available at https://github.com/BiQiWHU/CMD