As most recently proposed methods for human detection have achieved a sufficiently high recall rate within a reasonable number of proposals, in this paper, we mainly focus on how to improve the precision rate of human detectors. In order to address the two main challenges in precision improvement, i.e., i) hard background instances and ii) redundant partial proposals, we propose the novel PoseHD framework, a top-down pose-based approach on the basis of an arbitrary state-of-theart human detector. In our proposed PoseHD framework, we first make use of human pose estimation (in a batch manner) and present pose heatmap classification (by a convolutional neural network) to eliminate hard negatives by extracting the more detailed structural information; then, we utilize posebased proposal clustering and reranking modules, filtering redundant partial proposals by comprehensively considering (a) Positive instances (b) Hard negative instances both holistic and part information. The experimental results on multiple pedestrian benchmark datasets validate that our proposed PoseHD framework can generally improve the overall performance of recent state-of-the-art human detectors (by 2-4% in both mAP and MR metrics). Moreover, our PoseHD framework can be easily extended to object detection with large-scale object part annotations. Finally, in this paper, we present extensive ablative analysis to compare our approach with these traditional bottom-up pose-based models and highlight (c) Redundant partial proposals (in blue box) the importance of our framework design decisions.

Scene recognition remains one of the most challenging problems in image understanding. With the help of fully connected layers (FCL) and rectified linear units (ReLu), deep networks can extract the moderately sparse and discriminative feature representation required for scene recognition. However, few methods consider exploiting a sparsity model for learning the feature representation in order to provide enhanced discriminative capability. In this paper, we replace the conventional FCL and ReLu with a new dictionary learning layer, that is composed of a finite number of recurrent units to simultaneously enhance the sparse representation and discriminative abilities of features via the determination of optimal dictionaries. In addition, with the help of the structure of the dictionary, we propose a new label discriminative regressor to boost the discrimination ability. We also propose new constraints to prevent overfitting by incorporating the advantage of the Mahalanobis and Euclidean distances to balance the recognition accuracy and generalization performance. Our proposed approach is evaluated using various scene datasets and shows superior performance to many state-of-the-art approaches.

Person re-identification (re-ID) tasks aim to identify the same person in multiple images captured from non-overlapping camera views. Most previous re-ID studies have attempted to solve this problem through either representation learning or metric learning, or by combining both techniques. Representation learning relies on the latent factors or attributes of the data. In most of these works, the dimensionality of the factors/attributes has to be manually determined for each new dataset. Thus, this approach is not robust. Metric learning optimizes a metric across the dataset to measure similarity according to distance. However, choosing the optimal method for computing these distances is data dependent, and learning the appropriate metric relies on a sufficient number of pair-wise labels. To overcome these limitations, we propose a novel algorithm for person re-ID, called semi-supervised Bayesian attribute learning. We introduce an Indian Buffet Process to identify the priors of the latent attributes. The dimensionality of attributes factors is then automatically determined by nonparametric Bayesian learning. Meanwhile, unlike traditional distance metric learning, we propose a re-identification probability distribution to describe how likely it is that a pair of images contains the same person. This technique relies solely on the latent attributes of both images. Moreover, pair-wise labels that are not known can be estimated from pair-wise labels that are known, making this a robust approach for semi-supervised learning. Extensive experiments demonstrate the superior performance of our algorithm over several state-of-the-art algorithms on small-scale datasets and comparable performance on large-scale re-ID datasets.

In this paper, we present a Character-Aware Neural Network (Char-Net) for recognizing distorted scene text. Our Char-Net is composed of a word-level encoder, a character-level encoder, and a LSTM-based decoder. Unlike previous work which employed a global spatial transformer network to rectify the entire distorted text image, we take an approach of detecting and rectifying individual characters. To this end, we introduce a novel hierarchical attention mechanism (HAM) which consists of a recurrent RoIWarp layer and a character-level attention layer. The recurrent RoIWarp layer sequentially extracts a feature region corresponding to a character from the feature map produced by the word-level encoder, and feeds it to the character-level encoder which removes the distortion of the character through a simple spatial transformer and further encodes the character region. The character-level attention layer then attends to the most relevant features of the feature map produced by the character-level encoder and composes a context vector, which is finally fed to the LSTM-based decoder for decoding. This approach of adopting a simple local transformation to model the distortion of individual characters not only results in an improved efficiency, but can also handle different types of distortion that are hard, if not impossible, to be modelled by a single global transformation. Experiments have been conducted on six public benchmark datasets. Our results show that Char-Net can achieve state-of-the-art performance on all the benchmarks, especially on the IC-IST which contains scene text with large distortion. Code will be made available.

Video-based action recognition with deep neural networks has shown remarkable progress. However, most of the existing approaches are too computationally expensive due to the complex network architecture. To address these problems, we propose a new real-time action recognition architecture, called Temporal Convolutional 3D Network (T-C3D), which learns video action representations in a hierarchical multi-granularity manner. Specifically, we combine a residual 3D convolutional neural network which captures complementary information on the appearance of a single frame and the motion between consecutive frames with a new temporal encoding method to explore the temporal dynamics of the whole video. Thus heavy calculations are avoided when doing the inference, which enables the method to be capable of real-time processing. On two challenging benchmark datasets, UCF101 and HMDB51, our method is significantly better than state-of-the-art real-time methods by over 5.4% in terms of accuracy and 2 times faster in terms of inference speed (969 frames per second), demonstrating comparable recognition performance to the state-of-the-art methods. The source code for the complete system as well as the pre-trained models are publicly available at https://github.com/tc3d.

Action recognition is an important yet challenging task in computer vision. In this paper, we propose a novel deep-based framework for action recognition, which improves the recognition accuracy by: 1) deriving more precise features for representing actions, and 2) reducing the asynchrony between different information streams. We first introduce a coarse-to-fine network which extracts shared deep features at different action class granularities and progressively integrates them to obtain a more accurate feature representation for input actions. We further introduce an asynchronous fusion network. It fuses information from different streams by asynchronously integrating stream-wise features at different time points, hence better leveraging the complementary information in different streams. Experimental results on action recognition benchmarks demonstrate that our approach achieves the state-of-the-art performance.

Restoring face images from distortions is important in face recognition applications and is challenged by multiple scale issues, which is still not well-solved in research area. In this paper, we present a Sequential Gating Ensemble Network (SGEN) for multi-scale face restoration issue. We first employ the principle of ensemble learning into SGEN architecture design to reinforce predictive performance of the network. The SGEN aggregates multi-level base-encoders and base-decoders into the network, which enables the network to contain multiple scales of receptive field. Instead of combining these base-en/decoders directly with non-sequential operations, the SGEN takes base-en/decoders from different levels as sequential data. Specifically, the SGEN learns to sequentially extract high level information from base-encoders in bottom-up manner and restore low level information from base-decoders in top-down manner. Besides, we propose to realize bottom-up and top-down information combination and selection with Sequential Gating Unit (SGU). The SGU sequentially takes two inputs from different levels and decides the output based on one active input. Experiment results demonstrate that our SGEN is more effective at multi-scale human face restoration with more image details and less noise than state-of-the-art image restoration models. By using adversarial training, SGEN also produces more visually preferred results than other models through subjective evaluation.

Conventional methods of 3D object generative modeling learn volumetric predictions using deep networks with 3D convolutional operations, which are direct analogies to classical 2D ones. However, these methods are computationally wasteful in attempt to predict 3D shapes, where information is rich only on the surfaces. In this paper, we propose a novel 3D generative modeling framework to efficiently generate object shapes in the form of dense point clouds. We use 2D convolutional operations to predict the 3D structure from multiple viewpoints and jointly apply geometric reasoning with 2D projection optimization. We introduce the pseudo-renderer, a differentiable module to approximate the true rendering operation, to synthesize novel depth maps for optimization. Experimental results for single-image 3D object reconstruction tasks show that we outperforms state-of-the-art methods in terms of shape similarity and prediction density.

Visual relationship detection aims to describe the interactions between pairs of objects. Different from individual object learning tasks, the number of possible relationships are much larger, which makes it hard to explore only based on the visual appearance of objects. In addition, due to the limited human effort, the annotations for visual relationships are usually incomplete which increases the difficulty of model training and evaluation. In this paper, we propose a novel framework, called Deep Structural Ranking, for visual relationship detection. To complement the representation ability of visual appearance, we integrate multiple cues for predicting the relationships contained in an input image. Moreover, we design a new ranking objective function by enforcing the annotated relationships to have higher relevance scores. Unlike previous works, our proposed method can both facilitate the co-occurrence of relationships and mitigate the incompleteness problem. Experimental results show that our proposed method outperforms the state-of-the-art on the two widely used datasets. We also demonstrate its superiority in detecting zero-shot relationships.

Generating videos from text has proven to be a significant challenge for existing generative models. We tackle this problem by training a conditional generative model to extract both static and dynamic information from text. This is manifested in a hybrid framework, employing a Variational Autoencoder (VAE) and a Generative Adversarial Network (GAN). The static features, called "gist," are used to sketch text-conditioned background color and object layout structure. Dynamic features are considered by transforming input text into an image filter. To obtain a large amount of data for training the deep-learning model, we develop a method to automatically create a matched text-video corpus from publicly available online videos. Experimental results show that the proposed framework generates plausible and diverse short-duration smooth videos, while accurately reflecting the input text information. It significantly outperforms baseline models that directly adapt text-to-image generation procedures to produce videos. Performance is evaluated both visually and by adapting the inception score used to evaluate image generation in GANs.