Input binarization has shown to be an effective way for network acceleration. However, previous binarization scheme could be regarded as simple pixel-wise thresholding operations (i.e., order-one approximation) and suffers a big accuracy loss. In this paper, we propose a highorder binarization scheme, which achieves more accurate approximation while still possesses the advantage of binary operation. In particular, the proposed scheme recursively performs residual quantization and yields a series of binary input images with decreasing magnitude scales. Accordingly, we propose high-order binary filtering and gradient propagation operations for both forward and backward computations. Theoretical analysis shows approximation error guarantee property of proposed method. Extensive experimental results demonstrate that the proposed scheme yields great recognition accuracy while being accelerated.
Monte Carlo tree search (MCTS) is extremely popular in computer Go which determines each action by enormous simulations in a broad and deep search tree. However, human experts select most actions by pattern analysis and careful evaluation rather than brute search of millions of future nteractions. In this paper, we propose a computer Go system that follows experts way of thinking and playing. Our system consists of two parts. The first part is a novel deep alternative neural network (DANN) used to generate candidates of next move. Compared with existing deep convolutional neural network (DCNN), DANN inserts recurrent layer after each convolutional layer and stacks them in an alternative manner. We show such setting can preserve more contexts of local features and its evolutions which are beneficial for move prediction. The second part is a long-term evaluation (LTE) module used to provide a reliable evaluation of candidates rather than a single probability from move predictor. This is consistent with human experts nature of playing since they can foresee tens of steps to give an accurate estimation of candidates. In our system, for each candidate, LTE calculates a cumulative reward after several future interactions when local variations are settled. Combining criteria from the two parts, our system determines the optimal choice of next move. For more comprehensive experiments, we introduce a new professional Go dataset (PGD), consisting of 253233 professional records. Experiments on GoGoD and PGD datasets show the DANN can substantially improve performance of move prediction over pure DCNN. When combining LTE, our system outperforms most relevant approaches and open engines based on MCTS.
In this paper, we present a novel method of no-reference image quality assessment (NR-IQA), which is to predict the perceptual quality score of a given image without using any reference image. The proposed method harnesses three functions (i) the visual attention mechanism, which affects many aspects of visual perception including image quality assessment, however, is overlooked in the NR-IQA literature. The method assumes that the fixation areas on an image contain key information to the process of IQA. (ii) the robust averaging strategy, which is a means \--- supported by psychology studies \--- to integrating multiple/step-wise evidence to make a final perceptual judgment. (iii) the multi-task learning, which is believed to be an effectual means to shape representation learning and could result in a more generalized model. To exploit the synergy of the three, we consider the NR-IQA as a dynamic perception process, in which the model samples a sequence of "informative" areas and aggregates the information to learn a representation for the tasks of jointly predicting the image quality score and the distortion type. The model learning is implemented by a reinforcement strategy, in which the rewards of both tasks guide the learning of the optimal sampling policy to acquire the "task-informative" image regions so that the predictions can be made accurately and efficiently (in terms of the sampling steps). The reinforcement learning is realized by a deep network with the policy gradient method and trained through back-propagation. In experiments, the model is tested on the TID2008 dataset and it outperforms several state-of-the-art methods. Furthermore, the model is very efficient in the sense that a small number of fixations are used in NR-IQA.
This paper provides an overview of the on-going compact descriptors for video analysis standard (CDVA) from the ISO/IEC moving pictures experts group (MPEG). MPEG-CDVA targets at defining a standardized bitstream syntax to enable interoperability in the context of video analysis applications. During the developments of MPEGCDVA, a series of techniques aiming to reduce the descriptor size and improve the video representation ability have been proposed. This article describes the new standard that is being developed and reports the performance of these key technical contributions.
Dictionary learning is a challenge topic in many image processing areas. The basic goal is to learn a sparse representation from an overcomplete basis set. Due to combining the advantages of generic multiscale representations with learning based adaptivity, multiscale dictionary representation approaches have the power in capturing structural characteristics of natural images. However, existing multiscale learning approaches still suffer from three main weaknesses: inadaptability to diverse scales of image data, sensitivity to noise and outliers, difficulty to determine optimal dictionary structure. In this paper, we present a novel multiscale dictionary learning paradigm for sparse image representations based on an improved empirical mode decomposition. This powerful data-driven analysis tool for multi-dimensional signal can fully adaptively decompose the image into multiscale oscillating components according to intrinsic modes of data self. This treatment can obtain a robust and effective sparse representation, and meanwhile generates a raw base dictionary at multiple geometric scales and spatial frequency bands. This dictionary is refined by selecting optimal oscillating atoms based on frequency clustering. In order to further enhance sparsity and generalization, a tolerance dictionary is learned using a coherence regularized model. A fast proximal scheme is developed to optimize this model. The multiscale dictionary is considered as the product of oscillating dictionary and tolerance dictionary. Experimental results demonstrate that the proposed learning approach has the superior performance in sparse image representations as compared with several competing methods. We also show the promising results in image denoising application.
In this letter, we propose a novel image denoising method based on correlation preserving sparse coding. Because the instable and unreliable correlations among basis set can limit the performance of the dictionary-driven denoising methods, two effective regularized strategies are employed in the coding process. Specifically, a graph-based regularizer is built for preserving the global similarity correlations, which can adaptively capture both the geometrical structures and discriminative features of textured patches. In particular, edge weights in the graph are obtained by seeking a nonnegative low-rank construction. Besides, a robust locality-constrained coding can automatically preserve not only spatial neighborhood information but also internal consistency present in noisy patches while learning overcomplete dictionary. Experimental results demonstrate that our proposed method achieves state-of-the-art denoising performance in terms of both PSNR and subjective visual quality.
Aerial images are often degraded by space-varying motion blur and simultaneous uneven illumination. To recover high-quality aerial image from its non-uniform version, we propose a novel patch-wise restoration approach based on a key observation that the degree of blurring is inevitably affected by the illuminated conditions. A non-local Retinex model is developed to accurately estimate the reflectance component from the degraded aerial image. Thereafter the uneven illumination is corrected well. And then non-uniform coupled blurring in the enhanced reflectance image is alleviated and transformed towards uniform distribution, which will facilitate the subsequent deblurring. For constructing the multi-scale sparsified regularizer, the discrete shearlet transform is improved to better represent anisotropic image features in term of directional sensitivity and selectivity. In addition, a new adaptive variant of total generalized variation is proposed for the structure-preserving regularizer. These complementary regularizers are elegantly integrated into an objective function. The final deblurred image with uniform illumination can be extracted by applying the fast alternating direction scheme to solve the derived function. The experimental results demonstrate that our algorithm can not only remove both the space-varying illumination and motion blur in the aerial image effectively but also recover the abundant details of aerial scenes with top-level objective and subjective quality, and outperforms other state-of-the-art restoration methods.
The visual appearance of a person is easily affected by many factors like pose variations, viewpoint changes and camera parameter differences. This makes person Re-Identification (ReID) among multiple cameras a very challenging task. This work is motivated to learn mid-level human attributes which are robust to such visual appearance variations. And we propose a semi-supervised attribute learning framework which progressively boosts the accuracy of attributes only using a limited number of labeled data. Specifically, this framework involves a three-stage training. A deep Convolutional Neural Network (dCNN) is first trained on an independent dataset labeled with attributes. Then it is fine-tuned on another dataset only labeled with person IDs using our defined triplet loss. Finally, the updated dCNN predicts attribute labels for the target dataset, which is combined with the independent dataset for the final round of fine-tuning. The predicted attributes, namely \emph{deep attributes} exhibit superior generalization ability across different datasets. By directly using the deep attributes with simple Cosine distance, we have obtained surprisingly good accuracy on four person ReID datasets. Experiments also show that a simple metric learning modular further boosts our method, making it significantly outperform many recent works.
The iterations of many sparse estimation algorithms are comprised of a fixed linear filter cascaded with a thresholding nonlinearity, which collectively resemble a typical neural network layer. Consequently, a lengthy sequence of algorithm iterations can be viewed as a deep network with shared, hand-crafted layer weights. It is therefore quite natural to examine the degree to which a learned network model might act as a viable surrogate for traditional sparse estimation in domains where ample training data is available. While the possibility of a reduced computational budget is readily apparent when a ceiling is imposed on the number of layers, our work primarily focuses on estimation accuracy. In particular, it is well-known that when a signal dictionary has coherent columns, as quantified by a large RIP constant, then most tractable iterative algorithms are unable to find maximally sparse representations. In contrast, we demonstrate both theoretically and empirically the potential for a trained deep network to recover minimal $\ell_0$-norm representations in regimes where existing methods fail. The resulting system is deployed on a practical photometric stereo estimation problem, where the goal is to remove sparse outliers that can disrupt the estimation of surface normals from a 3D scene.
The pose problem is one of the bottlenecks in automatic face recognition. We argue that one of the diffculties in this problem is the severe misalignment in face images or feature vectors with different poses. In this paper, we propose that this problem can be statistically solved or at least mitigated by maximizing the intra-subject across-pose correlations via canonical correlation analysis (CCA). In our method, based on the data set with coupled face images of the same identities and across two different poses, CCA learns simultaneously two linear transforms, each for one pose. In the transformed subspace, the intra-subject correlations between the different poses are maximized, which implies pose-invariance or pose-robustness is achieved. The experimental results show that our approach could considerably improve the recognition performance. And if further enhanced with holistic+local feature representation, the performance could be comparable to the state-of-the-art.