Incomplete multi-view multi-label classification aims to accurately predict labels for each sample in the face of some missing views. Due to its widespread presence in real-world scenarios, it has become an extensively researched topic. In addition to the challenges brought by missing views, it also encounters issues caused by redundant views, whose inclusion fails to make a positive contribution to performance. In this paper, we make the first attempt to take advantage of diffusion models to address the missing view problem and design a strategy to identify and remove redundant views. Specifically, we train a diffusion model conditioned on the pseudo-labels to recover information of missing views. The learned diffusion model can carry data distribution knowledge in training split to the data. Regarding redundant identification strategy, it is designed by considering both the additional information of views and the classification difficulty level of samples, thereby adaptively identifying and removing redundant views. We conduct extensive experiments on five datasets, and the proposed method achieves favorable performance against several state-of-the-art methods on the multi-view multi-label classification task.

As a problem often encountered in real-world scenarios, multi-view multi-label learning has attracted considerable research attention. However, due to oversights in data col-lection and uncertainties in manual annotation, real-world data often suffer from incompleteness. Regrettably, most existing multi-view multi-label learning methods sidestep missing views and labels. Furthermore, they often neglect the potential of harnessing complementary information be-tween views and labels, thus constraining their classification capabilities. To address these challenges, we propose a view-category interactive sharing transformer tailored for incomplete multi-view multi-label learning. Within this net-work, we incorporate a two-layer transformer module to characterize the interplay between views and labels. Additionally, to address view incompleteness, a KNN-style missing view generation module is employed. Finally, we in-troduce a view-category consistency guided embedding en-hancement module to align different views and improve the discriminating power of the embeddings. Collectively, these modules synergistically integrate to classify the incomplete multi-view multi-label data effectively. Extensive experi-ments substantiate that our approach outperforms the ex-isting state-of-the-art methods.

One big challenge to federated learning is the non-IID data distribution caused by imbalanced classes. Existing federated learning approaches tend to bias towards classes containing a larger number of samples during local updates, which causes unwanted drift in the local classifiers. To address this issue, we propose a classifier debiased federated learning framework named FedCD for non-IID data. We introduce a novel hierarchical prototype contrastive learning strategy to learn fine-grained prototypes for each class. The prototypes characterize the sample distribution within each class, which helps align the features learned in the representation layer of every client's local model. At the representation layer, we use fine-grained prototypes to rebalance the class distribution on each client and rectify the classification layer of each local model. To alleviate the bias of the classification layer of the local models, we incorporate a global information distillation method to enable the local classifier to learn decoupled global classification information. We also adaptively aggregate the class-level classifiers based on their quality to reduce the impact of unreliable classes in each aggregated classifier. This mitigates the impact of client-side classifier bias on the global classifier. Comprehensive experiments conducted on various datasets show that our method, FedCD, effectively corrects classifier bias and outperforms state-of-the-art federated learning methods.

Siamese-based trackers currently are the dominant tracking paradigm due to the balance between speed and performance. However, it is prone to drift and tracking failure when the environment is complex and similar objects interfere. While the Siamese-based trackers perform the correlation operation, the responses of the target object and background appear in different channels, i.e., the feature spaces of the target object and background have some orthogonality. However, when meeting background clutters and similar objects interfere, this orthogonality becomes weaker and the wrong classification contribution of the object and the background reduces the stability of the learned similarity function, leading to many misclassified pixels in the heatmaps. In this work, we proposed a SiamORPN to solve the above issues. It is incorporated at two levels: an Orthogonal Region Proposal Network (ORPN) and an Adaptive Pixel-wise Aggregation (APA) module. Specifically, for ORPN, the orthogonality between the object and the background maximizes the inter-class inertia. Moreover, the ORPN introduces the orthogonal module to enhance this orthogonality. For APA, it introduces two lightweight networks to predict the weights of all pixels in different heatmaps and the weights of all pixels in different regression offsets. Experiments on challenging benchmarks, including OTB2015, VOT2016, VOT2018, GOT-10k test set, UAV123, LaSOT, and TrackingNet, demonstrate the proposed SiamORPN outperforms many SOTA trackers and achieves leading performance. The inference speed at GTX1080Ti can reach about 32 FPS, meeting the real-time requirements.