Hyperspectral change detection plays an essential role of monitoring the dynamic urban development and detecting precise fine object evolution and alteration. In this paper, we have proposed an original Efficient Multi-temporal Self-attention Network (EMS-Net) for hyperspectral change detection. The designed EMS module cuts redundancy of those similar and containing-no-changes feature maps, computing efficient multi-temporal change information for precise binary change map. Besides, to explore the clustering characteristics of the change detection, a novel supervised contrastive loss is provided to enhance the compactness of the unchanged. Experiments implemented on two hyperspectral change detection datasets manifests the out-standing performance and validity of proposed method.
Very-high-resolution (VHR) remote sensing (RS) image change detection (CD) has been a challenging task for its very rich spatial information and sample imbalance problem. In this paper, we have proposed a hierarchical change guiding map network (HCGMNet) for change detection. The model uses hierarchical convolution operations to extract multiscale features, continuously merges multi-scale features layer by layer to improve the expression of global and local information, and guides the model to gradually refine edge features and comprehensive performance by a change guide module (CGM), which is a self-attention with changing guide map. Extensive experiments on two CD datasets show that the proposed HCGMNet architecture achieves better CD performance than existing state-of-the-art (SOTA) CD methods.
In a federated learning (FL) system, malicious participants can easily embed backdoors into the aggregated model while maintaining the model's performance on the main task. To this end, various defenses, including training stage aggregation-based defenses and post-training mitigation defenses, have been proposed recently. While these defenses obtain reasonable performance against existing backdoor attacks, which are mainly heuristics based, we show that they are insufficient in the face of more advanced attacks. In particular, we propose a general reinforcement learning-based backdoor attack framework where the attacker first trains a (non-myopic) attack policy using a simulator built upon its local data and common knowledge on the FL system, which is then applied during actual FL training. Our attack framework is both adaptive and flexible and achieves strong attack performance and durability even under state-of-the-art defenses.
Multimodal headline utilizes both video frames and transcripts to generate the natural language title of the videos. Due to a lack of large-scale, manually annotated data, the task of annotating grounded headlines for video is labor intensive and impractical. Previous researches on pre-trained language models and video-language models have achieved significant progress in related downstream tasks. However, none of them can be directly applied to multimodal headline architecture where we need both multimodal encoder and sentence decoder. A major challenge in simply gluing language model and video-language model is the modality balance, which is aimed at combining visual-language complementary abilities. In this paper, we propose a novel approach to graft the video encoder from the pre-trained video-language model on the generative pre-trained language model. We also present a consensus fusion mechanism for the integration of different components, via inter/intra modality relation. Empirically, experiments show that the grafted model achieves strong results on a brand-new dataset collected from real-world applications.
Unsupervised multimodal change detection is a practical and challenging topic that can play an important role in time-sensitive emergency applications. To address the challenge that multimodal remote sensing images cannot be directly compared due to their modal heterogeneity, we take advantage of two types of modality-independent structural relationships in multimodal images. In particular, we present a structural relationship graph representation learning framework for measuring the similarity of the two structural relationships. Firstly, structural graphs are generated from preprocessed multimodal image pairs by means of an object-based image analysis approach. Then, a structural relationship graph convolutional autoencoder (SR-GCAE) is proposed to learn robust and representative features from graphs. Two loss functions aiming at reconstructing vertex information and edge information are presented to make the learned representations applicable for structural relationship similarity measurement. Subsequently, the similarity levels of two structural relationships are calculated from learned graph representations and two difference images are generated based on the similarity levels. After obtaining the difference images, an adaptive fusion strategy is presented to fuse the two difference images. Finally, a morphological filtering-based postprocessing approach is employed to refine the detection results. Experimental results on five datasets with different modal combinations demonstrate the effectiveness of the proposed method.
Neural ranking models (NRMs) have achieved promising results in information retrieval. NRMs have also been shown to be vulnerable to adversarial examples. A typical Word Substitution Ranking Attack (WSRA) against NRMs was proposed recently, in which an attacker promotes a target document in rankings by adding human-imperceptible perturbations to its text. This raises concerns when deploying NRMs in real-world applications. Therefore, it is important to develop techniques that defend against such attacks for NRMs. In empirical defenses adversarial examples are found during training and used to augment the training set. However, such methods offer no theoretical guarantee on the models' robustness and may eventually be broken by other sophisticated WSRAs. To escape this arms race, rigorous and provable certified defense methods for NRMs are needed. To this end, we first define the \textit{Certified Top-$K$ Robustness} for ranking models since users mainly care about the top ranked results in real-world scenarios. A ranking model is said to be Certified Top-$K$ Robust on a ranked list when it is guaranteed to keep documents that are out of the top $K$ away from the top $K$ under any attack. Then, we introduce a Certified Defense method, named CertDR, to achieve certified top-$K$ robustness against WSRA, based on the idea of randomized smoothing. Specifically, we first construct a smoothed ranker by applying random word substitutions on the documents, and then leverage the ranking property jointly with the statistical property of the ensemble to provably certify top-$K$ robustness. Extensive experiments on two representative web search datasets demonstrate that CertDR can significantly outperform state-of-the-art empirical defense methods for ranking models.
Image and language modeling is of crucial importance for vision-language pre-training (VLP), which aims to learn multi-modal representations from large-scale paired image-text data. However, we observe that most existing VLP methods focus on modeling the interactions between image and text features while neglecting the information disparity between image and text, thus suffering from focal bias. To address this problem, we propose a vision-language masked autoencoder framework (VLMAE). VLMAE employs visual generative learning, facilitating the model to acquire fine-grained and unbiased features. Unlike the previous works, VLMAE pays attention to almost all critical patches in an image, providing more comprehensive understanding. Extensive experiments demonstrate that VLMAE achieves better performance in various vision-language downstream tasks, including visual question answering, image-text retrieval and visual grounding, even with up to 20% pre-training speedup.
This technical report presents the 3rd winning solution for MTVG, a new task introduced in the 4-th Person in Context (PIC) Challenge at ACM MM 2022. MTVG aims at localizing the temporal boundary of the step in an untrimmed video based on a textual description. The biggest challenge of this task is the fi ne-grained video-text semantics of make-up steps. However, current methods mainly extract video features using action-based pre-trained models. As actions are more coarse-grained than make-up steps, action-based features are not sufficient to provide fi ne-grained cues. To address this issue,we propose to achieve fi ne-grained representation via exploiting feature diversities. Specifically, we proposed a series of methods from feature extraction, network optimization, to model ensemble. As a result, we achieved 3rd place in the MTVG competition.
The fast development of self-supervised learning lowers the bar learning feature representation from massive unlabeled data and has triggered a series of research on change detection of remote sensing images. Challenges in adapting self-supervised learning from natural images classification to remote sensing images change detection arise from difference between the two tasks. The learned patch-level feature representations are not satisfying for the pixel-level precise change detection. In this paper, we proposed a novel pixel-level self-supervised hyperspectral spatial-spectral understanding network (HyperNet) to accomplish pixel-wise feature representation for effective hyperspectral change detection. Concretely, not patches but the whole images are fed into the network and the multi-temporal spatial-spectral features are compared pixel by pixel. Instead of processing the two-dimensional imaging space and spectral response dimension in hybrid style, a powerful spatial-spectral attention module is put forward to explore the spatial correlation and discriminative spectral features of multi-temporal hyperspectral images (HSIs), separately. Only the positive samples at the same location of bi-temporal HSIs are created and forced to be aligned, aiming at learning the spectral difference-invariant features. Moreover, a new similarity loss function named focal cosine is proposed to solve the problem of imbalanced easy and hard positive samples comparison, where the weights of those hard samples are enlarged and highlighted to promote the network training. Six hyperspectral datasets have been adopted to test the validity and generalization of proposed HyperNet. The extensive experiments demonstrate the superiority of HyperNet over the state-of-the-art algorithms on downstream hyperspectral change detection tasks.