Network architecture plays a key role in the deep learning-based computer vision system. The widely-used convolutional neural network and transformer treat the image as a grid or sequence structure, which is not flexible to capture irregular and complex objects. In this paper, we propose to represent the image as a graph structure and introduce a new Vision GNN (ViG) architecture to extract graph-level feature for visual tasks. We first split the image to a number of patches which are viewed as nodes, and construct a graph by connecting the nearest neighbors. Based on the graph representation of images, we build our ViG model to transform and exchange information among all the nodes. ViG consists of two basic modules: Grapher module with graph convolution for aggregating and updating graph information, and FFN module with two linear layers for node feature transformation. Both isotropic and pyramid architectures of ViG are built with different model sizes. Extensive experiments on image recognition and object detection tasks demonstrate the superiority of our ViG architecture. We hope this pioneering study of GNN on general visual tasks will provide useful inspiration and experience for future research. The PyTroch code will be available at https://github.com/huawei-noah/CV-Backbones and the MindSpore code will be avaiable at https://gitee.com/mindspore/models.
Deep Neural Networks (DNNs) are vulnerable to the black-box adversarial attack that is highly transferable. This threat comes from the distribution gap between adversarial and clean samples in feature space of the target DNNs. In this paper, we use Deep Generative Networks (DGNs) with a novel training mechanism to eliminate the distribution gap. The trained DGNs align the distribution of adversarial samples with clean ones for the target DNNs by translating pixel values. Different from previous work, we propose a more effective pixel level training constraint to make this achievable, thus enhancing robustness on adversarial samples. Further, a class-aware feature-level constraint is formulated for integrated distribution alignment. Our approach is general and applicable to multiple tasks, including image classification, semantic segmentation, and object detection. We conduct extensive experiments on different datasets. Our strategy demonstrates its unique effectiveness and generality against black-box attacks.
Image forensics, aiming to ensure the authenticity of the image, has made great progress in dealing with common image manipulation such as copy-move, splicing, and inpainting in the past decades. However, only a few researchers pay attention to an emerging editing technique called image recoloring, which can manipulate the color values of an image to give it a new style. To prevent it from being used maliciously, the previous approaches address the conventional recoloring from the perspective of inter-channel correlation and illumination consistency. In this paper, we try to explore a solution from the perspective of the spatial correlation, which exhibits the generic detection capability for both conventional and deep learning-based recoloring. Through theoretical and numerical analysis, we find that the recoloring operation will inevitably destroy the spatial correlation between pixels, implying a new prior of statistical discriminability. Based on such fact, we generate a set of spatial correlation features and learn the informative representation from the set via a convolutional neural network. To train our network, we use three recoloring methods to generate a large-scale and high-quality data set. Extensive experimental results in two recoloring scenes demonstrate that the spatial correlation features are highly discriminative. Our method achieves the state-of-the-art detection accuracy on multiple benchmark datasets and exhibits well generalization for unknown types of recoloring methods.
Image retargeting changes the aspect ratio of images while aiming to preserve content and minimise noticeable distortion. Fast and high-quality methods are particularly relevant at present, due to the large variety of image and display aspect ratios. We propose a retargeting method that quantifies and limits warping distortions with the use of content-aware cropping. The pipeline of the proposed approach consists of the following steps. First, an importance map of a source image is generated using deep semantic segmentation and saliency detection models. Then, a preliminary warping mesh is computed using axis aligned deformations, enhanced with the use of a distortion measure to ensure low warping deformations. Finally, the retargeted image is produced using a content-aware cropping algorithm. In order to evaluate our method, we perform a user study based on the RetargetMe benchmark. Experimental analyses show that our method outperforms recent approaches, while running in a fraction of their execution time.
Since labeling medical image data is a costly and labor-intensive process, active learning has gained much popularity in the medical image segmentation domain in recent years. A variety of active learning strategies have been proposed in the literature, but their effectiveness is highly dependent on the dataset and training scenario. To facilitate the comparison of existing strategies and provide a baseline for evaluating novel strategies, we evaluate the performance of several well-known active learning strategies on three datasets from the Medical Segmentation Decathlon. Additionally, we consider a strided sampling strategy specifically tailored to 3D image data. We demonstrate that both random and strided sampling act as strong baselines and discuss the advantages and disadvantages of the studied methods. To allow other researchers to compare their work to our results, we provide an open-source framework for benchmarking active learning strategies on a variety of medical segmentation datasets.
Assessing the critical view of safety in laparoscopic cholecystectomy requires accurate identification and localization of key anatomical structures, reasoning about their geometric relationships to one another, and determining the quality of their exposure. In this work, we propose to capture each of these aspects by modeling the surgical scene with a disentangled latent scene graph representation, which we can then process using a graph neural network. Unlike previous approaches using graph representations, we explicitly encode in our graphs semantic information such as object locations and shapes, class probabilities and visual features. We also incorporate an auxiliary image reconstruction objective to help train the latent graph representations. We demonstrate the value of these components through comprehensive ablation studies and achieve state-of-the-art results for critical view of safety prediction across multiple experimental settings.
Indoor scenes exhibit significant appearance variations due to myriad interactions between arbitrarily diverse object shapes, spatially-changing materials, and complex lighting. Shadows, highlights, and inter-reflections caused by visible and invisible light sources require reasoning about long-range interactions for inverse rendering, which seeks to recover the components of image formation, namely, shape, material, and lighting. In this work, our intuition is that the long-range attention learned by transformer architectures is ideally suited to solve longstanding challenges in single-image inverse rendering. We demonstrate with a specific instantiation of a dense vision transformer, IRISformer, that excels at both single-task and multi-task reasoning required for inverse rendering. Specifically, we propose a transformer architecture to simultaneously estimate depths, normals, spatially-varying albedo, roughness and lighting from a single image of an indoor scene. Our extensive evaluations on benchmark datasets demonstrate state-of-the-art results on each of the above tasks, enabling applications like object insertion and material editing in a single unconstrained real image, with greater photorealism than prior works. Code and data are publicly released at https://github.com/ViLab-UCSD/IRISformer.
Occluded person re-identification (ReID) is a challenging problem due to contamination from occluders, and existing approaches address the issue with prior knowledge cues, eg human body key points, semantic segmentations and etc, which easily fails in the presents of heavy occlusion and other humans as occluders. In this paper, we propose a feature pruning and consolidation (FPC) framework to circumvent explicit human structure parse, which mainly consists of a sparse encoder, a global and local feature ranking module, and a feature consolidation decoder. Specifically, the sparse encoder drops less important image tokens (mostly related to background noise and occluders) solely according to correlation within the class token attention instead of relying on prior human shape information. Subsequently, the ranking stage relies on the preserved tokens produced by the sparse encoder to identify k-nearest neighbors from a pre-trained gallery memory by measuring the image and patch-level combined similarity. Finally, we use the feature consolidation module to compensate pruned features using identified neighbors for recovering essential information while disregarding disturbance from noise and occlusion. Experimental results demonstrate the effectiveness of our proposed framework on occluded, partial and holistic Re-ID datasets. In particular, our method outperforms state-of-the-art results by at least 8.6% mAP and 6.0% Rank-1 accuracy on the challenging Occluded-Duke dataset.
Isolating and controlling specific features in the outputs of generative models in a user-friendly way is a difficult and open-ended problem. We develop techniques that allow an oracle user to generate an image they are envisioning in their head by answering a sequence of relative queries of the form \textit{"do you prefer image $a$ or image $b$?"} Our framework consists of a Conditional VAE that uses the collected relative queries to partition the latent space into preference-relevant features and non-preference-relevant features. We then use the user's responses to relative queries to determine the preference-relevant features that correspond to their envisioned output image. Additionally, we develop techniques for modeling the uncertainty in images' predicted preference-relevant features, allowing our framework to generalize to scenarios in which the relative query training set contains noise.
By replacing the lens with a thin optical element, lensless imaging enables new applications and solutions beyond those supported by traditional camera design and post-processing, e.g. compact and lightweight form factors and visual privacy. The latter arises from the highly multiplexed measurements of lensless cameras, which require knowledge of the imaging system to recover a recognizable image. In this work, we exploit this unique multiplexing property: casting the optics as an encoder that produces learned embeddings directly at the camera sensor. We do so in the context of image classification, where we jointly optimize the encoder's parameters and those of an image classifier in an end-to-end fashion. Our experiments show that jointly learning the lensless optical encoder and the digital processing allows for lower resolution embeddings at the sensor, and hence better privacy as it is much harder to recover meaningful images from these measurements. Additional experiments show that such an optimization allows for lensless measurements that are more robust to typical real-world image transformations. While this work focuses on classification, the proposed programmable lensless camera and end-to-end optimization can be applied to other computational imaging tasks.