ICCV 2019 论文列表

Most previous image matting methods require a roughly-specificed trimap as input, and estimate fractional alpha values for all pixels that are in the unknown region of the trimap. In this paper, we argue that directly estimating the alpha matte from a coarse trimap is a major limitation of previous methods, as this practice tries to address two difficult and inherently different problems at the same time: identifying true blending pixels inside the trimap region, and estimate accurate alpha values for them. We propose AdaMatting, a new end-to-end matting framework that disentangles this problem into two sub-tasks: trimap adaptation and alpha estimation. Trimap adaptation is a pixel-wise classification problem that infers the global structure of the input image by identifying definite foreground, background, and semi-transparent image regions. Alpha estimation is a regression problem that calculates the opacity value of each blended pixel. Our method separately handles these two sub-tasks within a single deep convolutional neural network (CNN). Extensive experiments show that AdaMatting has additional structure awareness and trimap fault-tolerance. Our method achieves the state-of-the-art performance on Adobe Composition-1k dataset both qualitatively and quantitatively. It is also the current best-performing method on the alphamatting.com online evaluation for all commonly-used metrics.

Majority of the image set based face recognition methods use a generatively learned model for each person that is learned independently by ignoring the other persons in the gallery set. In contrast to these methods, this paper introduces a novel method that searches for discriminative convex models that best fit to an individual's face images but at the same time are as far as possible from the images of other persons in the gallery. We learn discriminative convex models for both affine and convex hulls of image sets. During testing, distances from the query set images to these models are computed efficiently by using simple matrix multiplications, and the query set is assigned to the person in the gallery whose image set is closest to the query images. The proposed method significantly outperforms other methods using generative convex models in terms of both accuracy and testing time, and achieves the state-of-the-art results on four of the five tested datasets. Especially, the accuracy improvement is significant on the challenging PaSC, COX and ESOGU video datasets.

Despite promising performance achieved by deep con- volutional neural networks for non-occluded pedestrian de- tection, it remains a great challenge to detect partially oc- cluded pedestrians. Compared with non-occluded pedes- trian examples, it is generally more difficult to distinguish occluded pedestrian examples from background in featue space due to the missing of occluded parts. In this paper, we propose a discriminative feature transformation which en- forces feature separability of pedestrian and non-pedestrian examples to handle occlusions for pedestrian detection. Specifically, in feature space it makes pedestrian exam- ples approach the centroid of easily classified non-occluded pedestrian examples and pushes non-pedestrian examples close to the centroid of easily classified non-pedestrian ex- amples. Such a feature transformation partially compen- sates the missing contribution of occluded parts in feature space, therefore improving the performance for occluded pedestrian detection. We implement our approach in the Fast R-CNN framework by adding one transformation net- work branch. We validate the proposed approach on two widely used pedestrian detection datasets: Caltech and CityPersons. Experimental results show that our approach achieves promising performance for both non-occluded and occluded pedestrian detection.

Person re-identification (Re-ID) has undergone a rapid development with the blooming of deep neural network. Most methods are very easily affected by target misalignment and background clutter in the training process. In this paper, we propose a simple yet effective feedforward attention network to address the two mentioned problems, in which a novel consistent attention regularizer and an improved triplet loss are designed to learn foreground attentive features for person Re-ID. Specifically, the consistent attention regularizer aims to keep the deduced foreground masks similar from the low-level, mid-level and high-level feature maps. As a result, the network will focus on the foreground regions at the lower layers, which is benefit to learn discriminative features from the foreground regions at the higher layers. Last but not least, the improved triplet loss is introduced to enhance the feature learning capability, which can jointly minimize the intra-class distance and maximize the inter-class distance in each triplet unit. Experimental results on the Market1501, DukeMTMC-reID and CUHK03 datasets have shown that our method outperforms most of the state-of-the-art approaches.

Editing 3D models is a very challenging task, as it requires complex interactions with the 3D shape to reach the targeted design, while preserving the global consistency and plausibility of the shape. In this work, we present an intelligent and user-friendly 3D editing tool, where the edited model is constrained to lie onto a learned manifold of realistic shapes. Due to the topological variability of real 3D models, they often lie close to a disconnected manifold, which cannot be learned with a common learning algorithm. Therefore, our tool is based on a new deep learning model, DiscoNet, which extends 3D surface autoencoders in two ways. Firstly, our deep learning model uses several autoencoders to automatically learn each connected component of a disconnected manifold, without any supervision. Secondly, each autoencoder infers the output 3D surface by deforming a pre-learned 3D template specific to each connected component. Both advances translate into improved 3D synthesis, thus enhancing the quality of our 3D editing tool.

Per-pixel ground-truth depth data is challenging to acquire at scale. To overcome this limitation, self-supervised learning has emerged as a promising alternative for training models to perform monocular depth estimation. In this paper, we propose a set of improvements, which together result in both quantitatively and qualitatively improved depth maps compared to competing self-supervised methods. Research on self-supervised monocular training usually explores increasingly complex architectures, loss functions, and image formation models, all of which have recently helped to close the gap with fully-supervised methods. We show that a surprisingly simple model, and associated design choices, lead to superior predictions. In particular, we propose (i) a minimum reprojection loss, designed to robustly handle occlusions, (ii) a full-resolution multi-scale sampling method that reduces visual artifacts, and (iii) an auto-masking loss to ignore training pixels that violate camera motion assumptions. We demonstrate the effectiveness of each component in isolation, and show high quality, state-of-the-art results on the KITTI benchmark.

Hardware-friendly network quantization (e.g., binary/uniform quantization) can efficiently accelerate the inference and meanwhile reduce memory consumption of the deep neural networks, which is crucial for model deployment on resource-limited devices like mobile phones. However, due to the discreteness of low-bit quantization, existing quantization methods often face the unstable training process and severe performance degradation. To address this problem, in this paper we propose Differentiable Soft Quantization (DSQ) to bridge the gap between the full-precision and low-bit networks. DSQ can automatically evolve during training to gradually approximate the standard quantization. Owing to its differentiable property, DSQ can help pursue the accurate gradients in backward propagation, and reduce the quantization loss in forward process with an appropriate clipping range. Extensive experiments over several popular network structures show that training low-bit neural networks with DSQ can consistently outperform state-of-the-art quantization methods. Besides, our first efficient implementation for deploying 2 to 4-bit DSQ on devices with ARM architecture achieves up to 1.7x speed up, compared with the open-source 8-bit high-performance inference framework NCNN [31].

Group convolution, which divides the channels of ConvNets into groups, has achieved impressive improvement over the regular convolution operation. However, existing models, e.g. ResNext, still suffers from the sub-optimal performance due to manually defining the number of groups as a constant over all of the layers. Toward addressing this issue, we present Groupable ConvNet (GroupNet) built by using a novel dynamic grouping convolution (DGConv) operation, which is able to learn the number of groups in an end-to-end manner. The proposed approach has several appealing benefits. (1) DGConv provides a unified convolution representation and covers many existing convolution operations such as regular dense convolution, group convolution, and depthwise convolution. (2) DGConv is a differentiable and flexible operation which learns to perform various convolutions from training data. (3) GroupNet trained with DGConv learns different number of groups for different convolution layers. Extensive experiments demonstrate that GroupNet outperforms its counterparts such as ResNet and ResNeXt in terms of accuracy and computational complexity. We also present introspection and reproducibility study, for the first time, showing the learning dynamics of training group numbers.

This paper proposes a differentiable kernel evolution (DKE) algorithm to find a better layer-operator for the convolutional neural network. Unlike most of the other neural architecture searching (NAS) technologies, we consider the searching space in a fundamental scope: kernel space, which encodes the assembly of basic multiply-accumulate (MAC) operations into a conv-kernel. We first deduce a strict form of the generalized convolutional operator by some necessary constraints and construct a continuous searching space for its extra freedom-of-degree, namely, the connection of each MAC. Then a novel unsupervised greedy evolution algorithm called gradient agreement guided searching (GAGS) is proposed to learn the optimal location for each MAC in the spatially continuous searching space. We leverage DKE on multiple kinds of tasks such as object classification, face/object detection, large-scale fine-grained and recognition, with various kinds of backbone architecture. Not to mention the consistent performance gain, we found the proposed DKE can further act as an auto-dilated operator, which makes it easy to boost the performance of miniaturized neural networks in multiple tasks.

Capturing document images with hand-held devices in unstructured environments is a common practice nowadays. However, "casual" photos of documents are usually unsuitable for automatic information extraction, mainly due to physical distortion of the document paper, as well as various camera positions and illumination conditions. In this work, we propose DewarpNet, a deep-learning approach for document image unwarping from a single image. Our insight is that the 3D geometry of the document not only determines the warping of its texture but also causes the illumination effects. Therefore, our novelty resides on the explicit modeling of 3D shape for document paper in an end-to-end pipeline. Also, we contribute the largest and most comprehensive dataset for document image unwarping to date - Doc3D. This dataset features multiple ground-truth annotations, including 3D shape, surface normals, UV map, albedo image, etc. Training with Doc3D, we demonstrate state-of-the-art performance for DewarpNet with extensive qualitative and quantitative evaluations. Our network also significantly improves OCR performance on captured document images, decreasing character error rate by 42% on average. Both the code and the dataset are released.

Classical semantic segmentation methods, including the recent deep learning ones, assume that all classes observed at test time have been seen during training. In this paper, we tackle the more realistic scenario where unexpected objects of unknown classes can appear at test time. The main trends in this area either leverage the notion of prediction uncertainty to flag the regions with low confidence as unknown, or rely on autoencoders and highlight poorly-decoded regions. Having observed that, in both cases, the detected regions typically do not correspond to unexpected objects, in this paper, we introduce a drastically different strategy: It relies on the intuition that the network will produce spurious labels in regions depicting unexpected objects. Therefore, resynthesizing the image from the resulting semantic map will yield significant appearance differences with respect to the input image. In other words, we translate the problem of detecting unknown classes to one of identifying poorly-resynthesized image regions. We show that this outperforms both uncertainty- and autoencoder-based methods.

We seek to detect visual relations in images of the form of triplets t = (subject, predicate, object), such as "person riding dog", where training examples of the individual entities are available but their combinations are unseen at training. This is an important set-up due to the combinatorial nature of visual relations : collecting sufficient training data for all possible triplets would be very hard. The contributions of this work are three-fold. First, we learn a representation of visual relations that combines (i) individual embeddings for subject, object and predicate together with (ii) a visual phrase embedding that represents the relation triplet. Second, we learn how to transfer visual phrase embeddings from existing training triplets to unseen test triplets using analogies between relations that involve similar objects. Third, we demonstrate the benefits of our approach on three challenging datasets : on HICO-DET, our model achieves significant improvement over a strong baseline for both frequent and unseen triplets, and we observe similar improvement for the retrieval of unseen triplets with out-of-vocabulary predicates on the COCO-a dataset as well as the challenging unusual triplets in the UnRel dataset.

Most malicious photo manipulations are created using standard image editing tools, such as Adobe Photoshop. We present a method for detecting one very popular Photoshop manipulation -- image warping applied to human faces -- using a model trained entirely using fake images that were automatically generated by scripting Photoshop itself. We show that our model outperforms humans at the task of recognizing manipulated images, can predict the specific location of edits, and in some cases can be used to "undo" a manipulation to reconstruct the original, unedited image. We demonstrate that the system can be successfully applied to artist-created image manipulations.

Recent advances in deep learning greatly boost the performance of object detection. State-of-the-art methods such as Faster-RCNN, FPN and R-FCN have achieved high accuracy in challenging benchmark datasets. However, these methods require fully annotated object bounding boxes for training, which are incredibly hard to scale up due to the high annotation cost. Weakly-supervised methods, on the other hand, only require image-level labels for training, but the performance is far below their fully-supervised counterparts. In this paper, we propose a semi-supervised large scale fine-grained detection method, which only needs bounding box annotations of a smaller number of coarse-grained classes and image-level labels of large scale fine-grained classes, and can detect all classes at nearly fully-supervised accuracy. We achieve this by utilizing the correlations between coarse-grained and fine-grained classes with shared backbone, soft-attention based proposal re-ranking, and a dual-level memory module. Experiment results show that our methods can achieve close accuracy on object detection to state-of-the-art fully-supervised methods on two large scale datasets, ImageNet and OpenImages, with only a small fraction of fully annotated classes.

In this work, we propose a novel depth-induced multi-scale recurrent attention network for saliency detection. It achieves dramatic performance especially in complex scenarios. There are three main contributions of our network that are experimentally demonstrated to have significant practical merits. First, we design an effective depth refinement block using residual connections to fully extract and fuse multi-level paired complementary cues from RGB and depth streams. Second, depth cues with abundant spatial information are innovatively combined with multi-scale context features for accurately locating salient objects. Third, we boost our model's performance by a novel recurrent attention module inspired by Internal Generative Mechanism of human brain. This module can generate more accurate saliency results via comprehensively learning the internal semantic relation of the fused feature and progressively optimizing local details with memory-oriented scene understanding. In addition, we create a large scale RGB-D dataset containing more complex scenarios, which can contribute to comprehensively evaluating saliency models. Extensive experiments on six public datasets and ours demonstrate that our method can accurately identify salient objects and achieve consistently superior performance over 16 state-of-the-art RGB and RGB-D approaches.

We present a novel method for simultaneous learning of depth, egomotion, object motion, and camera intrinsics from monocular videos, using only consistency across neighboring video frames as supervision signal. Similarly to prior work, our method learns by applying differentiable warping to frames and comparing the result to adjacent ones, but it provides several improvements: We address occlusions geometrically and differentiably, directly using the depth maps as predicted during training. We introduce randomized layer normalization, a novel powerful regularizer, and we account for object motion relative to the scene. To the best of our knowledge, our work is the first to learn the camera intrinsic parameters, including lens distortion, from video in an unsupervised manner, thereby allowing us to extract accurate depth and motion from arbitrary videos of unknown origin at scale. We evaluate our results on the Cityscapes, KITTI and EuRoC datasets, establishing new state of the art on depth prediction and odometry, and demonstrate qualitatively that depth prediction can be learned from a collection of YouTube videos. The code will be open sourced once anonymity is lifted.

Depth completion aims to recover dense depth maps from sparse depth measurements. It is of increasing importance for autonomous driving and draws increasing attention from the vision community. Most of the current competitive methods directly train a network to learn a mapping from sparse depth inputs to dense depth maps, which has difficulties in utilizing the 3D geometric constraints and handling the practical sensor noises. In this paper, to regularize the depth completion and improve the robustness against noise, we propose a unified CNN framework that 1) models the geometric constraints between depth and surface normal in a diffusion module and 2) predicts the confidence of sparse LiDAR measurements to mitigate the impact of noise. Specifically, our encoder-decoder backbone predicts the surface normal, coarse depth and confidence of LiDAR inputs simultaneously, which are subsequently inputted into our diffusion refinement module to obtain the final completion results. Extensive experiments on KITTI depth completion dataset and NYU-Depth-V2 dataset demonstrate that our method achieves state-of-the-art performance. Further ablation study and analysis give more insights into the proposed components and demonstrate the generalization capability and stability of our model.

Point cloud processing is very challenging, as the diverse shapes formed by irregular points are often indistinguishable. A thorough grasp of the elusive shape requires sufficiently contextual semantic information, yet few works devote to this. Here we propose DensePoint, a general architecture to learn densely contextual representation for point cloud processing. Technically, it extends regular grid CNN to irregular point configuration by generalizing a convolution operator, which holds the permutation invariance of points, and achieves efficient inductive learning of local patterns. Architecturally, it finds inspiration from dense connection mode, to repeatedly aggregate multi-level and multi-scale semantics in a deep hierarchy. As a result, densely contextual information along with rich semantics, can be acquired by DensePoint in an organic manner, making it highly effective. Extensive experiments on challenging benchmarks across four tasks, as well as thorough model analysis, verify DensePoint achieves the state of the arts.

Object detection from images captured by Unmanned Aerial Vehicles (UAVs) is becoming increasingly useful. Despite the great success of the generic object detection methods trained on ground-to-ground images, a huge performance drop is observed when they are directly applied to images captured by UAVs. The unsatisfactory performance is owing to many UAV-specific nuisances, such as varying flying altitudes, adverse weather conditions, dynamically changing viewing angles, etc. Those nuisances constitute a large number of fine-grained domains, across which the detection model has to stay robust. Fortunately, UAVs will record meta-data that depict those varying attributes, which are either freely available along with the UAV images, or can be easily obtained. We propose to utilize those free meta-data in conjunction with associated UAV images to learn domain-robust features via an adversarial training framework dubbed Nuisance Disentangled Feature Transform (NDFT), for the specific challenging problem of object detection in UAV images, achieving a substantial gain in robustness to those nuisances. We demonstrate the effectiveness of our proposed algorithm, by showing state-of-the- art performance (single model) on two existing UAV-based object detection benchmarks. The code is available at https://github.com/TAMU-VITA/UAV-NDFT.