Class Activation Map (CAM) has emerged as a popular tool for weakly supervised semantic segmentation (WSSS), allowing the localization of object regions in an image using only image-level labels. However, existing CAM methods suffer from under-activation of target object regions and false-activation of background regions due to the fact that a lack of detailed supervision can hinder the model's ability to understand the image as a whole. In this paper, we propose a novel Question-Answer Cross-Language-Image Matching framework for WSSS (QA-CLIMS), leveraging the vision-language foundation model to maximize the text-based understanding of images and guide the generation of activation maps. First, a series of carefully designed questions are posed to the VQA (Visual Question Answering) model with Question-Answer Prompt Engineering (QAPE) to generate a corpus of both foreground target objects and backgrounds that are adaptive to query images. We then employ contrastive learning in a Region Image Text Contrastive (RITC) network to compare the obtained foreground and background regions with the generated corpus. Our approach exploits the rich textual information from the open vocabulary as additional supervision, enabling the model to generate high-quality CAMs with a more complete object region and reduce false-activation of background regions. We conduct extensive analysis to validate the proposed method and show that our approach performs state-of-the-art on both PASCAL VOC 2012 and MS COCO datasets. Code is available at: https://github.com/CVI-SZU/QA-CLIMS
Despite the remarkable success of existing methods for few-shot segmentation, there remain two crucial challenges. First, the feature learning for novel classes is suppressed during the training on base classes in that the novel classes are always treated as background. Thus, the semantics of novel classes are not well learned. Second, most of existing methods fail to consider the underlying semantic gap between the support and the query resulting from the representative bias by the scarce support samples. To circumvent these two challenges, we propose to activate the discriminability of novel classes explicitly in both the feature encoding stage and the prediction stage for segmentation. In the feature encoding stage, we design the Semantic-Preserving Feature Learning module (SPFL) to first exploit and then retain the latent semantics contained in the whole input image, especially those in the background that belong to novel classes. In the prediction stage for segmentation, we learn an Self-Refined Online Foreground-Background classifier (SROFB), which is able to refine itself using the high-confidence pixels of query image to facilitate its adaptation to the query image and bridge the support-query semantic gap. Extensive experiments on PASCAL-5$^i$ and COCO-20$^i$ datasets demonstrates the advantages of these two novel designs both quantitatively and qualitatively.
With the rapid increase of micro-video creators and viewers, how to make personalized recommendations from a large number of candidates to viewers begins to attract more and more attention. However, existing micro-video recommendation models rely on expensive multi-modal information and learn an overall interest embedding that cannot reflect the user's multiple interests in micro-videos. Recently, contrastive learning provides a new opportunity for refining the existing recommendation techniques. Therefore, in this paper, we propose to extract contrastive multi-interests and devise a micro-video recommendation model CMI. Specifically, CMI learns multiple interest embeddings for each user from his/her historical interaction sequence, in which the implicit orthogonal micro-video categories are used to decouple multiple user interests. Moreover, it establishes the contrastive multi-interest loss to improve the robustness of interest embeddings and the performance of recommendations. The results of experiments on two micro-video datasets demonstrate that CMI achieves state-of-the-art performance over existing baselines.
In this work, we present a fully self-supervised framework for semantic segmentation(FS^4). A fully bootstrapped strategy for semantic segmentation, which saves efforts for the huge amount of annotation, is crucial for building customized models from end-to-end for open-world domains. This application is eagerly needed in realistic scenarios. Even though recent self-supervised semantic segmentation methods have gained great progress, these works however heavily depend on the fully-supervised pretrained model and make it impossible a fully self-supervised pipeline. To solve this problem, we proposed a bootstrapped training scheme for semantic segmentation, which fully leveraged the global semantic knowledge for self-supervision with our proposed PGG strategy and CAE module. In particular, we perform pixel clustering and assignments for segmentation supervision. Preventing it from clustering a mess, we proposed 1) a pyramid-global-guided (PGG) training strategy to supervise the learning with pyramid image/patch-level pseudo labels, which are generated by grouping the unsupervised features. The stable global and pyramid semantic pseudo labels can prevent the segmentation from learning too many clutter regions or degrading to one background region; 2) in addition, we proposed context-aware embedding (CAE) module to generate global feature embedding in view of its neighbors close both in space and appearance in a non-trivial way. We evaluate our method on the large-scale COCO-Stuff dataset and achieved 7.19 mIoU improvements on both things and stuff objects
Graph Neural Networks (GNNs) have received increasing attention for representation learning in various machine learning tasks. However, most existing GNNs applying neighborhood aggregation usually perform poorly on the graph with heterophily where adjacent nodes belong to different classes. In this paper, we show that in typical heterphilous graphs, the edges may be directed, and whether to treat the edges as is or simply make them undirected greatly affects the performance of the GNN models. Furthermore, due to the limitation of heterophily, it is highly beneficial for the nodes to aggregate messages from similar nodes beyond local neighborhood.These motivate us to develop a model that adaptively learns the directionality of the graph, and exploits the underlying long-distance correlations between nodes. We first generalize the graph Laplacian to digraph based on the proposed Feature-Aware PageRank algorithm, which simultaneously considers the graph directionality and long-distance feature similarity between nodes. Then digraph Laplacian defines a graph propagation matrix that leads to a model called {\em DiglacianGCN}. Based on this, we further leverage the node proximity measured by commute times between nodes, in order to preserve the nodes' long-distance correlation on the topology level. Extensive experiments on ten datasets with different levels of homophily demonstrate the effectiveness of our method over existing solutions in the task of node classification.
The prevalence of relation networks in computer vision is in stark contrast to underexplored point-based methods. In this paper, we explore the possibilities of local relation operators and survey their feasibility. We propose a scalable and efficient module, called group relation aggregator. The module computes a feature of a group based on the aggregation of the features of the inner-group points weighted by geometric relations and semantic relations. We adopt this module to design our RPNet. We further verify the expandability of RPNet, in terms of both depth and width, on the tasks of classification and segmentation. Surprisingly, empirical results show that wider RPNet fits for classification, while deeper RPNet works better on segmentation. RPNet achieves state-of-the-art for classification and segmentation on challenging benchmarks. We also compare our local aggregator with PointNet++, with around 30% parameters and 50% computation saving. Finally, we conduct experiments to reveal the robustness of RPNet with regard to rigid transformation and noises.
Interactions between users and videos are the major data source of performing video recommendation. Despite lots of existing recommendation methods, user behaviors on videos, which imply the complex relations between users and videos, are still far from being fully explored. In the paper, we present a model named Sagittarius. Sagittarius adopts a graph convolutional neural network to capture the influence between users and videos. In particular, Sagittarius differentiates between different user behaviors by weighting and fuses the semantics of user behaviors into the embeddings of users and videos. Moreover, Sagittarius combines multiple optimization objectives to learn user and video embeddings and then achieves the video recommendation by the learned user and video embeddings. The experimental results on multiple datasets show that Sagittarius outperforms several state-of-the-art models in terms of recall, unique recall and NDCG.
In this work, we propose a novel straightforward method for medical volume and sequence segmentation with limited annotations. To avert laborious annotating, the recent success of self-supervised learning(SSL) motivates the pre-training on unlabeled data. Despite its success, it is still challenging to adapt typical SSL methods to volume/sequence segmentation, due to their lack of mining on local semantic discrimination and rare exploitation on volume and sequence structures. Based on the continuity between slices/frames and the common spatial layout of organs across volumes/sequences, we introduced a novel bootstrap self-supervised representation learning method by leveraging the predictable possibility of neighboring slices. At the core of our method is a simple and straightforward dense self-supervision on the predictions of local representations and a strategy of predicting locals based on global context, which enables stable and reliable supervision for both global and local representation mining among volumes. Specifically, we first proposed an asymmetric network with an attention-guided predictor to enforce distance-specific prediction and supervision on slices within and across volumes/sequences. Secondly, we introduced a novel prototype-based foreground-background calibration module to enhance representation consistency. The two parts are trained jointly on labeled and unlabeled data. When evaluated on three benchmark datasets of medical volumes and sequences, our model outperforms existing methods with a large margin of 4.5\% DSC on ACDC, 1.7\% on Prostate, and 2.3\% on CAMUS. Intensive evaluations reveals the effectiveness and superiority of our method.
In this paper, we investigate if we could make the self-training -- a simple but popular framework -- work better for semi-supervised segmentation. Since the core issue in semi-supervised setting lies in effective and efficient utilization of unlabeled data, we notice that increasing the diversity and hardness of unlabeled data is crucial to performance improvement. Being aware of this fact, we propose to adopt the most plain self-training scheme coupled with appropriate strong data augmentations on unlabeled data (namely ST) for this task, which surprisingly outperforms previous methods under various settings without any bells and whistles. Moreover, to alleviate the negative impact of the wrongly pseudo labeled images, we further propose an advanced self-training framework (namely ST++), that performs selective re-training via selecting and prioritizing the more reliable unlabeled images. As a result, the proposed ST++ boosts the performance of semi-supervised model significantly and surpasses existing methods by a large margin on the Pascal VOC 2012 and Cityscapes benchmark. Overall, we hope this straightforward and simple framework will serve as a strong baseline or competitor for future works. Code is available at https://github.com/LiheYoung/ST-PlusPlus.
We consider graph representation learning in a self-supervised manner. Graph neural networks (GNNs) use neighborhood aggregation as a core component that results in feature smoothing among nodes in proximity. While successful in various prediction tasks, such a paradigm falls short of capturing nodes' similarities over a long distance, which proves to be important for high-quality learning. To tackle this problem, we strengthen the graph with two additional graph views, in which nodes are directly linked to those with the most similar features or local structures. Not restricted by connectivity in the original graph, the generated views allow the model to enhance its expressive power with new and complementary perspectives from which to look at the relationship between nodes. Following a contrastive learning approach, We propose a method that aims to maximize the agreement between representations across generated views and the original graph. We also propose a channel-level contrast approach that greatly reduces computation cost, compared to the commonly used node level contrast, which requires computation cost quadratic in the number of nodes. Extensive experiments on seven assortative graphs and four disassortative graphs demonstrate the effectiveness of our approach.