Vision and Language Navigation (VLN) requires an agent to navigate to a target location by following natural language instructions. Most of existing works represent a navigation candidate by the feature of the corresponding single view where the candidate lies in. However, an instruction may mention landmarks out of the single view as references, which might lead to failures of textual-visual matching of existing methods. In this work, we propose a multi-module Neighbor-View Enhanced Model (NvEM) to adaptively incorporate visual contexts from neighbor views for better textual-visual matching. Specifically, our NvEM utilizes a subject module and a reference module to collect contexts from neighbor views. The subject module fuses neighbor views at a global level, and the reference module fuses neighbor objects at a local level. Subjects and references are adaptively determined via attention mechanisms. Our model also includes an action module to utilize the strong orientation guidance (e.g., ``turn left'') in instructions. Each module predicts navigation action separately and their weighted sum is used for predicting the final action. Extensive experimental results demonstrate the effectiveness of the proposed method on the R2R and R4R benchmarks against several state-of-the-art navigators, and NvEM even beats some pre-training ones. Our code is available at https://github.com/MarSaKi/NvEM.
Transfer learning with pre-training on large-scale datasets has played an increasingly significant role in computer vision and natural language processing recently. However, as there exist numerous application scenarios that have distinctive demands such as certain latency constraints and specialized data distributions, it is prohibitively expensive to take advantage of large-scale pre-training for per-task requirements. In this paper, we focus on the area of object detection and present a transfer learning system named GAIA, which could automatically and efficiently give birth to customized solutions according to heterogeneous downstream needs. GAIA is capable of providing powerful pre-trained weights, selecting models that conform to downstream demands such as latency constraints and specified data domains, and collecting relevant data for practitioners who have very few datapoints for their tasks. With GAIA, we achieve promising results on COCO, Objects365, Open Images, Caltech, CityPersons, and UODB which is a collection of datasets including KITTI, VOC, WiderFace, DOTA, Clipart, Comic, and more. Taking COCO as an example, GAIA is able to efficiently produce models covering a wide range of latency from 16ms to 53ms, and yields AP from 38.2 to 46.5 without whistles and bells. To benefit every practitioner in the community of object detection, GAIA is released at https://github.com/GAIA-vision.
Deep neural networks have shown their vulnerability to adversarial attacks. In this paper, we focus on sparse adversarial attack based on the $\ell_0$ norm constraint, which can succeed by only modifying a few pixels of an image. Despite a high attack success rate, prior sparse attack methods achieve a low transferability under the black-box protocol due to overfitting the target model. Therefore, we introduce a generator architecture to alleviate the overfitting issue and thus efficiently craft transferable sparse adversarial examples. Specifically, the generator decouples the sparse perturbation into amplitude and position components. We carefully design a random quantization operator to optimize these two components jointly in an end-to-end way. The experiment shows that our method has improved the transferability by a large margin under a similar sparsity setting compared with state-of-the-art methods. Moreover, our method achieves superior inference speed, 700$\times$ faster than other optimization-based methods. The code is available at https://github.com/shaguopohuaizhe/TSAA.
Few-shot learning is a challenging task since only few instances are given for recognizing an unseen class. One way to alleviate this problem is to acquire a strong inductive bias via meta-learning on similar tasks. In this paper, we show that such inductive bias can be learned from a flat collection of unlabeled images, and instantiated as transferable representations among seen and unseen classes. Specifically, we propose a novel part-based self-supervised representation learning scheme to learn transferable representations by maximizing the similarity of an image to its discriminative part. To mitigate the overfitting in few-shot classification caused by data scarcity, we further propose a part augmentation strategy by retrieving extra images from a base dataset. We conduct systematic studies on miniImageNet and tieredImageNet benchmarks. Remarkably, our method yields impressive results, outperforming the previous best unsupervised methods by 7.74% and 9.24% under 5-way 1-shot and 5-way 5-shot settings, which are comparable with state-of-the-art supervised methods.
Previous methods decompose the blind super-resolution (SR) problem into two sequential steps: \textit{i}) estimating the blur kernel from given low-resolution (LR) image and \textit{ii}) restoring the SR image based on the estimated kernel. This two-step solution involves two independently trained models, which may not be well compatible with each other. A small estimation error of the first step could cause a severe performance drop of the second one. While on the other hand, the first step can only utilize limited information from the LR image, which makes it difficult to predict a highly accurate blur kernel. Towards these issues, instead of considering these two steps separately, we adopt an alternating optimization algorithm, which can estimate the blur kernel and restore the SR image in a single model. Specifically, we design two convolutional neural modules, namely \textit{Restorer} and \textit{Estimator}. \textit{Restorer} restores the SR image based on the predicted kernel, and \textit{Estimator} estimates the blur kernel with the help of the restored SR image. We alternate these two modules repeatedly and unfold this process to form an end-to-end trainable network. In this way, \textit{Estimator} utilizes information from both LR and SR images, which makes the estimation of the blur kernel easier. More importantly, \textit{Restorer} is trained with the kernel estimated by \textit{Estimator}, instead of the ground-truth kernel, thus \textit{Restorer} could be more tolerant to the estimation error of \textit{Estimator}. Extensive experiments on synthetic datasets and real-world images show that our model can largely outperform state-of-the-art methods and produce more visually favorable results at a much higher speed. The source code is available at \url{https://github.com/greatlog/DAN.git}.
Maliciously-manipulated images or videos - so-called deep fakes - especially face-swap images and videos have attracted more and more malicious attackers to discredit some key figures. Previous pixel-level artifacts based detection techniques always focus on some unclear patterns but ignore some available semantic clues. Therefore, these approaches show weak interpretability and robustness. In this paper, we propose a biometric information based method to fully exploit the appearance and shape feature for face-swap detection of key figures. The key aspect of our method is obtaining the inconsistency of 3D facial shape and facial appearance, and the inconsistency based clue offers natural interpretability for the proposed face-swap detection method. Experimental results show the superiority of our method in robustness on various laundering and cross-domain data, which validates the effectiveness of the proposed method.
In recent years, virtual makeup applications have become more and more popular. However, it is still challenging to propose a robust makeup transfer method in the real-world environment. Current makeup transfer methods mostly work well on good-conditioned clean makeup images, but transferring makeup that exhibits shadow and occlusion is not satisfying. To alleviate it, we propose a novel makeup transfer method, called 3D-Aware Shadow and Occlusion Robust GAN (SOGAN). Given the source and the reference faces, we first fit a 3D face model and then disentangle the faces into shape and texture. In the texture branch, we map the texture to the UV space and design a UV texture generator to transfer the makeup. Since human faces are symmetrical in the UV space, we can conveniently remove the undesired shadow and occlusion from the reference image by carefully designing a Flip Attention Module (FAM). After obtaining cleaner makeup features from the reference image, a Makeup Transfer Module (MTM) is introduced to perform accurate makeup transfer. The qualitative and quantitative experiments demonstrate that our SOGAN not only achieves superior results in shadow and occlusion situations but also performs well in large pose and expression variations.
Referring image segmentation aims to segment the objects referred by a natural language expression. Previous methods usually focus on designing an implicit and recurrent feature interaction mechanism to fuse the visual-linguistic features to directly generate the final segmentation mask without explicitly modeling the localization information of the referent instances. To tackle these problems, we view this task from another perspective by decoupling it into a "Locate-Then-Segment" (LTS) scheme. Given a language expression, people generally first perform attention to the corresponding target image regions, then generate a fine segmentation mask about the object based on its context. The LTS first extracts and fuses both visual and textual features to get a cross-modal representation, then applies a cross-model interaction on the visual-textual features to locate the referred object with position prior, and finally generates the segmentation result with a light-weight segmentation network. Our LTS is simple but surprisingly effective. On three popular benchmark datasets, the LTS outperforms all the previous state-of-the-art methods by a large margin (e.g., +3.2% on RefCOCO+ and +3.4% on RefCOCOg). In addition, our model is more interpretable with explicitly locating the object, which is also proved by visualization experiments. We believe this framework is promising to serve as a strong baseline for referring image segmentation.
Generalizable person Re-Identification (ReID) has attracted growing attention in recent computer vision community, as it offers ready-to-use ReID models without the need for model retraining in new environments. In this work, we introduce causality into person ReID and propose a novel generalizable framework, named Domain Invariant Representations for generalizable person Re-Identification (DIR-ReID). We assume the data generation process is controlled by two sets of factors, i.e. identity-specific factors containing identity related cues, and domain-specific factors describing other scene-related information which cause distribution shifts across domains. With the assumption above, a novel Multi-Domain Disentangled Adversarial Network (MDDAN) is designed to disentangle these two sets of factors. Furthermore, a Causal Data Augmentation (CDA) block is proposed to perform feature-level data augmentation for better domain-invariant representations, which can be explained as interventions on latent factors from a causal learning perspective. Extensive experiments have been conducted, showing that DIR-ReID outperforms state-of-the-art methods on large-scale domain generalization (DG) ReID benchmarks. Moreover, a theoretical analysis is provided for a better understanding of our method.
Graph classification is a challenging research problem in many applications across a broad range of domains. In these applications, it is very common that class distribution is imbalanced. Recently, Graph Neural Network (GNN) models have achieved superior performance on various real-world datasets. Despite their success, most of current GNN models largely overlook the important setting of imbalanced class distribution, which typically results in prediction bias towards majority classes. To alleviate the prediction bias, we propose to leverage semantic structure of dataset based on the distribution of node embedding. Specifically, we present GraphDIVE, a general framework leveraging mixture of diverse experts (i.e., graph classifiers) for imbalanced graph classification. With a divide-and-conquer principle, GraphDIVE employs a gating network to partition an imbalanced graph dataset into several subsets. Then each expert network is trained based on its corresponding subset. Experiments on real-world imbalanced graph datasets demonstrate the effectiveness of GraphDIVE.