Wen Li

Jiale Tao, Biao Wang, Borun Xu et al.

Given a source image and a driving video depicting the same object type, the motion transfer task aims to generate a video by learning the motion from the driving video while preserving the appearance from the source image. In this paper, we propose a novel structure-aware motion modeling approach, the deformable anchor model (DAM), which can automatically discover the motion structure of arbitrary objects without leveraging their prior structure information. Specifically, inspired by the known deformable part model (DPM), our DAM introduces two types of anchors or keypoints: i) a number of motion anchors that capture both appearance and motion information from the source image and driving video; ii) a latent root anchor, which is linked to the motion anchors to facilitate better learning of the representations of the object structure information. Moreover, DAM can be further extended to a hierarchical version through the introduction of additional latent anchors to model more complicated structures. By regularizing motion anchors with latent anchor(s), DAM enforces the correspondences between them to ensure the structural information is well captured and preserved. Moreover, DAM can be learned effectively in an unsupervised manner. We validate our proposed DAM for motion transfer on different benchmark datasets. Extensive experiments clearly demonstrate that DAM achieves superior performance relative to existing state-of-the-art methods.

Borun Xu, Biao Wang, Jinhong Deng et al.

Motion transfer aims to transfer the motion of a driving video to a source image. When there are considerable differences between object in the driving video and that in the source image, traditional single domain motion transfer approaches often produce notable artifacts; for example, the synthesized image may fail to preserve the human shape of the source image (cf . Fig. 1 (a)). To address this issue, in this work, we propose a Motion and Appearance Adaptation (MAA) approach for cross-domain motion transfer, in which we regularize the object in the synthesized image to capture the motion of the object in the driving frame, while still preserving the shape and appearance of the object in the source image. On one hand, considering the object shapes of the synthesized image and the driving frame might be different, we design a shape-invariant motion adaptation module that enforces the consistency of the angles of object parts in two images to capture the motion information. On the other hand, we introduce a structure-guided appearance consistency module designed to regularize the similarity between the corresponding patches of the synthesized image and the source image without affecting the learned motion in the synthesized image. Our proposed MAA model can be trained in an end-to-end manner with a cyclic reconstruction loss, and ultimately produces a satisfactory motion transfer result (cf . Fig. 1 (b)). We conduct extensive experiments on human dancing dataset Mixamo-Video to Fashion-Video and human face dataset Vox-Celeb to Cufs; on both of these, our MAA model outperforms existing methods both quantitatively and qualitatively.

Fanyue Wei, Biao Wang, Tiezheng Ge et al.

The goal of video highlight detection is to select the most attractive segments from a long video to depict the most interesting parts of the video. Existing methods typically focus on modeling relationship between different video segments in order to learning a model that can assign highlight scores to these segments; however, these approaches do not explicitly consider the contextual dependency within individual segments. To this end, we propose to learn pixel-level distinctions to improve the video highlight detection. This pixel-level distinction indicates whether or not each pixel in one video belongs to an interesting section. The advantages of modeling such fine-level distinctions are two-fold. First, it allows us to exploit the temporal and spatial relations of the content in one video, since the distinction of a pixel in one frame is highly dependent on both the content before this frame and the content around this pixel in this frame. Second, learning the pixel-level distinction also gives a good explanation to the video highlight task regarding what contents in a highlight segment will be attractive to people. We design an encoder-decoder network to estimate the pixel-level distinction, in which we leverage the 3D convolutional neural networks to exploit the temporal context information, and further take advantage of the visual saliency to model the spatial distinction. State-of-the-art performance on three public benchmarks clearly validates the effectiveness of our framework for video highlight detection.

Linlong Fan, Ye Huang, Yanqi Ge et al.

Existing view-based methods excel at recognizing 3D objects from predefined viewpoints, but their exploration of recognition under arbitrary views is limited. This is a challenging and realistic setting because each object has different viewpoint positions and quantities, and their poses are not aligned. However, most view-based methods, which aggregate multiple view features to obtain a global feature representation, hard to address 3D object recognition under arbitrary views. Due to the unaligned inputs from arbitrary views, it is challenging to robustly aggregate features, leading to performance degradation. In this paper, we introduce a novel Part-aware Network (PANet), which is a part-based representation, to address these issues. This part-based representation aims to localize and understand different parts of 3D objects, such as airplane wings and tails. It has properties such as viewpoint invariance and rotation robustness, which give it an advantage in addressing the 3D object recognition problem under arbitrary views. Our results on benchmark datasets clearly demonstrate that our proposed method outperforms existing view-based aggregation baselines for the task of 3D object recognition under arbitrary views, even surpassing most fixed viewpoint methods.

Yawei Li, Wen Li, Martin Danelljan et al.

In this paper, we tackle the problem of convolutional neural network design. Instead of focusing on the design of the overall architecture, we investigate a design space that is usually overlooked, i.e. adjusting the channel configurations of predefined networks. We find that this adjustment can be achieved by shrinking widened baseline networks and leads to superior performance. Based on that, we articulate the heterogeneity hypothesis: with the same training protocol, there exists a layer-wise differentiated network architecture (LW-DNA) that can outperform the original network with regular channel configurations but with a lower level of model complexity. The LW-DNA models are identified without extra computational cost or training time compared with the original network. This constraint leads to controlled experiments which direct the focus to the importance of layer-wise specific channel configurations. LW-DNA models come with advantages related to overfitting, i.e. the relative relationship between model complexity and dataset size. Experiments are conducted on various networks and datasets for image classification, visual tracking and image restoration. The resultant LW-DNA models consistently outperform the baseline models. Code is available at https://github.com/ofsoundof/Heterogeneity_Hypothesis.

Xingyu Chen, Junzhi Yu, Shihan Kong et al.

Object detection has been vigorously investigated for years but fast accurate detection for real-world scenes remains a very challenging problem. Overcoming drawbacks of single-stage detectors, we take aim at precisely detecting objects for static and temporal scenes in real time. Firstly, as a dual refinement mechanism, a novel anchor-offset detection is designed, which includes an anchor refinement, a feature location refinement, and a deformable detection head. This new detection mode is able to simultaneously perform two-step regression and capture accurate object features. Based on the anchor-offset detection, a dual refinement network (DRNet) is developed for high-performance static detection, where a multi-deformable head is further designed to leverage contextual information for describing objects. As for temporal detection in videos, temporal refinement networks (TRNet) and temporal dual refinement networks (TDRNet) are developed by propagating the refinement information across time. We also propose a soft refinement strategy to temporally match object motion with the previous refinement. Our proposed methods are evaluated on PASCAL VOC, COCO, and ImageNet VID datasets. Extensive comparisons on static and temporal detection verify the superiority of DRNet, TRNet, and TDRNet. Consequently, our developed approaches run in a fairly fast speed, and in the meantime achieve a significantly enhanced detection accuracy, i.e., 84.4% mAP on VOC 2007, 83.6% mAP on VOC 2012, 69.4% mAP on VID 2017, and 42.4% AP on COCO. Ultimately, producing encouraging results, our methods are applied to online underwater object detection and grasping with an autonomous system. Codes are publicly available at https://github.com/SeanChenxy/TDRN.

Weichen Zhang, Dong Xu, Wanli Ouyang et al.

This paper proposes a new unsupervised domain adaptation approach called Collaborative and Adversarial Network (CAN), which uses the domain-collaborative and domain-adversarial learning strategy for training the neural network. The domain-collaborative learning aims to learn domain-specific feature representation to preserve the discriminability for the target domain, while the domain adversarial learning aims to learn domain-invariant feature representation to reduce the domain distribution mismatch between the source and target domains. We show that these two learning strategies can be uniformly formulated as domain classifier learning with positive or negative weights on the losses. We then design a collaborative and adversarial training scheme, which automatically learns domain-specific representations from lower blocks in CNNs through collaborative learning and domain-invariant representations from higher blocks through adversarial learning. Moreover, to further enhance the discriminability in the target domain, we propose Self-Paced CAN (SPCAN), which progressively selects pseudo-labeled target samples for re-training the classifiers. We employ a self-paced learning strategy to select pseudo-labeled target samples in an easy-to-hard fashion. Comprehensive experiments on different benchmark datasets, Office-31, ImageCLEF-DA, and VISDA-2017 for the object recognition task, and UCF101-10 and HMDB51-10 for the video action recognition task, show our newly proposed approaches achieve the state-of-the-art performance, which clearly demonstrates the effectiveness of our proposed approaches for unsupervised domain adaptation.

Yin Zhang, Jinhong Deng, Peidong Liu et al.

Visual detection of Micro Air Vehicles (MAVs) has attracted increasing attention in recent years due to its important application in various tasks. The existing methods for MAV detection assume that the training set and testing set have the same distribution. As a result, when deployed in new domains, the detectors would have a significant performance degradation due to domain discrepancy. In this paper, we study the problem of cross-domain MAV detection. The contributions of this paper are threefold. 1) We propose a Multi-MAV-Multi-Domain (M3D) dataset consisting of both simulation and realistic images. Compared to other existing datasets, the proposed one is more comprehensive in the sense that it covers rich scenes, diverse MAV types, and various viewing angles. A new benchmark for cross-domain MAV detection is proposed based on the proposed dataset. 2) We propose a Noise Suppression Network (NSN) based on the framework of pseudo-labeling and a large-to-small training procedure. To reduce the challenging pseudo-label noises, two novel modules are designed in this network. The first is a prior-based curriculum learning module for allocating adaptive thresholds for pseudo labels with different difficulties. The second is a masked copy-paste augmentation module for pasting truly-labeled MAVs on unlabeled target images and thus decreasing pseudo-label noises. 3) Extensive experimental results verify the superior performance of the proposed method compared to the state-of-the-art ones. In particular, it achieves mAP of 46.9%(+5.8%), 50.5%(+3.7%), and 61.5%(+11.3%) on the tasks of simulation-to-real adaptation, cross-scene adaptation, and cross-camera adaptation, respectively.

Wen Li, Furong Xu, Jianan Zhao et al.

Triplet loss is a widely adopted loss function in ReID task which pulls the hardest positive pairs close and pushes the hardest negative pairs far away. However, the selected samples are not the hardest globally, but the hardest only in a mini-batch, which will affect the performance. In this report, a hard batch mining method is proposed to mine the hardest samples globally to make triplet harder. More specifically, the most similar classes are selected into a same mini-batch so that the similar classes could be pushed further away. Besides, an adversarial scene removal module composed of a scene classifier and an adversarial loss is used to learn scene invariant feature representations. Experiments are conducted on dataset MSMT17 to prove the effectiveness, and our method surpasses all of the previous methods and sets state-of-the-art result.

Jiayi Tian, Jing Zhang, Wen Li et al.

Domain adaptation aims to leverage a label-rich domain (the source domain) to help model learning in a label-scarce domain (the target domain). Most domain adaptation methods require the co-existence of source and target domain samples to reduce the distribution mismatch, however, access to the source domain samples may not always be feasible in the real world applications due to different problems (e.g., storage, transmission, and privacy issues). In this work, we deal with the source data-free unsupervised domain adaptation problem, and propose a novel approach referred to as Virtual Domain Modeling (VDM-DA). The virtual domain acts as a bridge between the source and target domains. On one hand, we generate virtual domain samples based on an approximated Gaussian Mixture Model (GMM) in the feature space with the pre-trained source model, such that the virtual domain maintains a similar distribution with the source domain without accessing to the original source data. On the other hand, we also design an effective distribution alignment method to reduce the distribution divergence between the virtual domain and the target domain by gradually improving the compactness of the target domain distribution through model learning. In this way, we successfully achieve the goal of distribution alignment between the source and target domains by training deep networks without accessing to the source domain data. We conduct extensive experiments on benchmark datasets for both 2D image-based and 3D point cloud-based cross-domain object recognition tasks, where the proposed method referred to Domain Adaptation with Virtual Domain Modeling (VDM-DA) achieves the state-of-the-art performances on all datasets.

Rui Gong, Dengxin Dai, Yuhua Chen et al.

Image-to-image translation is to map images from a given \emph{style} to another given \emph{style}. While exceptionally successful, current methods assume the availability of training images in both source and target domains, which does not always hold in practice. Inspired by humans' reasoning capability of analogy, we propose analogical image translation (AIT). Given images of two styles in the source domain: $\mathcal{A}$ and $\mathcal{A}^\prime$, along with images $\mathcal{B}$ of the first style in the target domain, learn a model to translate $\mathcal{B}$ to $\mathcal{B}^\prime$ in the target domain, such that $\mathcal{A}:\mathcal{A}^\prime ::\mathcal{B}:\mathcal{B}^\prime$. AIT is especially useful for translation scenarios in which training data of one style is hard to obtain but training data of the same two styles in another domain is available. For instance, in the case from normal conditions to extreme, rare conditions, obtaining real training images for the latter case is challenging but obtaining synthetic data for both cases is relatively easy. In this work, we are interested in adding adverse weather effects, more specifically fog effects, to images taken in clear weather. To circumvent the challenge of collecting real foggy images, AIT learns with synthetic clear-weather images, synthetic foggy images and real clear-weather images to add fog effects onto real clear-weather images without seeing any real foggy images during training. AIT achieves this zero-shot image translation capability by coupling a supervised training scheme in the synthetic domain, a cycle consistency strategy in the real domain, an adversarial training scheme between the two domains, and a novel network design. Experiments show the effectiveness of our method for zero-short image translation and its benefit for downstream tasks such as semantic foggy scene understanding.

Yuhua Chen, Wen Li, Xiaoran Chen et al.

Recently, increasing attention has been drawn to training semantic segmentation models using synthetic data and computer-generated annotation. However, domain gap remains a major barrier and prevents models learned from synthetic data from generalizing well to real-world applications. In this work, we take the advantage of additional geometric information from synthetic data, a powerful yet largely neglected cue, to bridge the domain gap. Such geometric information can be generated easily from synthetic data, and is proven to be closely coupled with semantic information. With the geometric information, we propose a model to reduce domain shift on two levels: on the input level, we augment the traditional image translation network with the additional geometric information to translate synthetic images into realistic styles; on the output level, we build a task network which simultaneously performs depth estimation and semantic segmentation on the synthetic data. Meanwhile, we encourage the network to preserve correlation between depth and semantics by adversarial training on the output space. We then validate our method on two pairs of synthetic to real dataset: Virtual KITTI to KITTI, and SYNTHIA to Cityscapes, where we achieve a significant performance gain compared to the non-adapt baseline and methods using only semantic label. This demonstrates the usefulness of geometric information from synthetic data for cross-domain semantic segmentation.

Hai Hu, Wen Li, Sandra Kübler

We present a machine learning approach to distinguish texts translated to Chinese (by humans) from texts originally written in Chinese, with a focus on a wide range of syntactic features. Using Support Vector Machines (SVMs) as classifier on a genre-balanced corpus in translation studies of Chinese, we find that constituent parse trees and dependency triples as features without lexical information perform very well on the task, with an F-measure above 90%, close to the results of lexical n-gram features, without the risk of learning topic information rather than translation features. Thus, we claim syntactic features alone can accurately distinguish translated from original Chinese. Translated Chinese exhibits an increased use of determiners, subject position pronouns, NP + 'de' as NP modifiers, multiple NPs or VPs conjoined by a Chinese specific punctuation, among other structures. We also interpret the syntactic features with reference to previous translation studies in Chinese, particularly the usage of pronouns.

Dengxin Dai, Wen Li, Till Kroeger et al.

Manifold theory has been the central concept of many learning methods. However, learning modern CNNs with manifold structures has not raised due attention, mainly because of the inconvenience of imposing manifold structures onto the architecture of the CNNs. In this paper we present ManifoldNet, a novel method to encourage learning of manifold-aware representations. Our approach segments the input manifold into a set of fragments. By assigning the corresponding segmentation id as a pseudo label to every sample, we convert the problem of preserving the local manifold structure into a point-wise classification task. Due to its unsupervised nature, the segmentation tends to be noisy. We mitigate this by introducing ensemble manifold segmentation (EMS). EMS accounts for the manifold structure by dividing the training data into an ensemble of classification training sets that contain samples of local proximity. CNNs are trained on these ensembles under a multi-task learning framework to conform to the manifold. ManifoldNet can be trained with only the pseudo labels or together with task-specific labels. We evaluate ManifoldNet on two different tasks: network imitation (distillation) and semi-supervised learning. Our experiments show that the manifold structures are effectively utilized for both unsupervised and semi-supervised learning.

Yuhua Chen, Wen Li, Christos Sakaridis et al.

Object detection typically assumes that training and test data are drawn from an identical distribution, which, however, does not always hold in practice. Such a distribution mismatch will lead to a significant performance drop. In this work, we aim to improve the cross-domain robustness of object detection. We tackle the domain shift on two levels: 1) the image-level shift, such as image style, illumination, etc, and 2) the instance-level shift, such as object appearance, size, etc. We build our approach based on the recent state-of-the-art Faster R-CNN model, and design two domain adaptation components, on image level and instance level, to reduce the domain discrepancy. The two domain adaptation components are based on H-divergence theory, and are implemented by learning a domain classifier in adversarial training manner. The domain classifiers on different levels are further reinforced with a consistency regularization to learn a domain-invariant region proposal network (RPN) in the Faster R-CNN model. We evaluate our newly proposed approach using multiple datasets including Cityscapes, KITTI, SIM10K, etc. The results demonstrate the effectiveness of our proposed approach for robust object detection in various domain shift scenarios.

Yuhua Chen, Wen Li, Luc Van Gool

Exploiting synthetic data to learn deep models has attracted increasing attention in recent years. However, the intrinsic domain difference between synthetic and real images usually causes a significant performance drop when applying the learned model to real world scenarios. This is mainly due to two reasons: 1) the model overfits to synthetic images, making the convolutional filters incompetent to extract informative representation for real images; 2) there is a distribution difference between synthetic and real data, which is also known as the domain adaptation problem. To this end, we propose a new reality oriented adaptation approach for urban scene semantic segmentation by learning from synthetic data. First, we propose a target guided distillation approach to learn the real image style, which is achieved by training the segmentation model to imitate a pretrained real style model using real images. Second, we further take advantage of the intrinsic spatial structure presented in urban scene images, and propose a spatial-aware adaptation scheme to effectively align the distribution of two domains. These two modules can be readily integrated with existing state-of-the-art semantic segmentation networks to improve their generalizability when adapting from synthetic to real urban scenes. We evaluate the proposed method on Cityscapes dataset by adapting from GTAV and SYNTHIA datasets, where the results demonstrate the effectiveness of our method.

Limin Wang, Wei Li, Wen Li et al.

Spatiotemporal feature learning in videos is a fundamental problem in computer vision. This paper presents a new architecture, termed as Appearance-and-Relation Network (ARTNet), to learn video representation in an end-to-end manner. ARTNets are constructed by stacking multiple generic building blocks, called as SMART, whose goal is to simultaneously model appearance and relation from RGB input in a separate and explicit manner. Specifically, SMART blocks decouple the spatiotemporal learning module into an appearance branch for spatial modeling and a relation branch for temporal modeling. The appearance branch is implemented based on the linear combination of pixels or filter responses in each frame, while the relation branch is designed based on the multiplicative interactions between pixels or filter responses across multiple frames. We perform experiments on three action recognition benchmarks: Kinetics, UCF101, and HMDB51, demonstrating that SMART blocks obtain an evident improvement over 3D convolutions for spatiotemporal feature learning. Under the same training setting, ARTNets achieve superior performance on these three datasets to the existing state-of-the-art methods.

Yifei Wang, Wen Li, Dengxin Dai et al.

In this paper, we propose a new approach called Deep LogCORAL for unsupervised visual domain adaptation. Our work builds on the recently proposed Deep CORAL method, which proposed to train a convolutional neural network and simultaneously minimize the Euclidean distance of convariance matrices between the source and target domains. We propose to use the Riemannian distance, approximated by Log-Euclidean distance, to replace the naive Euclidean distance in Deep CORAL. We also consider first-order information, and minimize the distance of mean vectors between two domains. We build an end-to-end model, in which we minimize both the classification loss, and the domain difference based on the first and second order information between two domains. Our experiments on the benchmark Office dataset demonstrate the improvements of our newly proposed Deep LogCORAL approach over the Deep CORAL method, as well as further improvement when optimizing both orders of information.

Wen Li, Limin Wang, Wei Li et al.

We present the 2017 WebVision Challenge, a public image recognition challenge designed for deep learning based on web images without instance-level human annotation. Following the spirit of previous vision challenges, such as ILSVRC, Places2 and PASCAL VOC, which have played critical roles in the development of computer vision by contributing to the community with large scale annotated data for model designing and standardized benchmarking, we contribute with this challenge a large scale web images dataset, and a public competition with a workshop co-located with CVPR 2017. The WebVision dataset contains more than $2.4$ million web images crawled from the Internet by using queries generated from the $1,000$ semantic concepts of the benchmark ILSVRC 2012 dataset. Meta information is also included. A validation set and test set containing human annotated images are also provided to facilitate algorithmic development. The 2017 WebVision challenge consists of two tracks, the image classification task on WebVision test set, and the transfer learning task on PASCAL VOC 2012 dataset. In this paper, we describe the details of data collection and annotation, highlight the characteristics of the dataset, and introduce the evaluation metrics.

Muhammad Ghifary, W. Bastiaan Kleijn, Mengjie Zhang et al.

In this paper, we propose a novel unsupervised domain adaptation algorithm based on deep learning for visual object recognition. Specifically, we design a new model called Deep Reconstruction-Classification Network (DRCN), which jointly learns a shared encoding representation for two tasks: i) supervised classification of labeled source data, and ii) unsupervised reconstruction of unlabeled target data.In this way, the learnt representation not only preserves discriminability, but also encodes useful information from the target domain. Our new DRCN model can be optimized by using backpropagation similarly as the standard neural networks. We evaluate the performance of DRCN on a series of cross-domain object recognition tasks, where DRCN provides a considerable improvement (up to ~8% in accuracy) over the prior state-of-the-art algorithms. Interestingly, we also observe that the reconstruction pipeline of DRCN transforms images from the source domain into images whose appearance resembles the target dataset. This suggests that DRCN's performance is due to constructing a single composite representation that encodes information about both the structure of target images and the classification of source images. Finally, we provide a formal analysis to justify the algorithm's objective in domain adaptation context.