Channel attention mechanisms, as the key components of some modern convolutional neural networks (CNNs) architectures, have been commonly used in many visual tasks for effective performance improvement. It is able to reinforce the informative channels and to suppress useless channels of feature maps obtained by CNNs. Recently, different attention modules have been proposed, which are implemented in various ways. However, they are mainly based on convolution and pooling operations, which are lack of intuitive and reasonable insights about the principles that they are based on. Moreover, the ways that they improve the performance of the CNNs is not clear either. In this paper, we propose a Gaussian process embedded channel attention (GPCA) module and interpret the channel attention intuitively and reasonably in a probabilistic way. The GPCA module is able to model the correlations from channels which are assumed as beta distributed variables with Gaussian process prior. As the beta distribution is intractably integrated into the end-to-end training of the CNNs, we utilize an appropriate approximation of the beta distribution to make the distribution assumption implemented easily. In this case, the proposed GPCA module can be integrated into the end-to-end training of the CNNs. Experimental results demonstrate that the proposed GPCA module can improve the accuracies of image classification on four widely used datasets.
In this paper, we propose a dual-attention guided dropblock module, and target at learning the complementary and discriminative visual patterns for weakly supervised object localization (WSOL). We extend the attention mechanism in the task of WSOL, and carefully design two types of attention modules to capture the informative features for better feature representations. Based on two types of attention mechanism, we propose a channel attention guided dropout (CAGD) and a spatial attention guided dropblock (SAGD). The CAGD ranks channel attention according to a measure of importance and treat the top-k largest magnitude attentions as important ones. The SAGD can not only completely remove the information by erasing the contiguous regions of feature maps rather than individual pixels, but also simply sense the foreground objects and background regions to alleviate the attention misdirection. Extensive experiments demonstrate that the proposed method achieves new state-of-the-art localization accuracy on three challenging datasets.
The topic of object detection has been largely improved recently, especially with the development of convolutional neural network. However, there still exist a lot of challenging cases, such as small object, compact and dense or highly overlapping object. Existing methods can detect multiple objects wonderfully, but because of the slight changes between frames, the detection effect of the model will become unstable, the detection results may result in dropping or increasing the object. In the pedestrian flow detection task, such phenomenon can not accurately calculate the flow. To solve this problem, in this paper, we describe the new function for real-time multi-object detection in sports competition and pedestrians flow detection in public based on deep learning. Our work is to extract a video clip and solve this frame of clips efficiently. More specfically, our algorithm includes two stages: judge method and optimization method. The judge can set a maximum threshold for better results under the model, the threshold value corresponds to the upper limit of the algorithm with better detection results. The optimization method to solve detection jitter problem. Because of the occurrence of frame hopping in the video, and it will result in the generation of video fragments discontinuity. We use optimization algorithm to get the key value, and then the detection result value of index is replaced by key value to stabilize the change of detection result sequence. Based on the proposed algorithm, we adopt wanfang sports competition dataset as the main test dataset and our own test dataset for YOLOv3-Abnormal Number Version(YOLOv3-ANV), which is 5.4% average improvement compared with existing methods. Also, video above the threshold value can be obtained for further analysis. Spontaneously, our work also can used for pedestrians flow detection and pedestrian alarm tasks.
Existing sketch-analysis work studies sketches depicting static objects or scenes. In this work, we propose a novel cross-modal retrieval problem of fine-grained instance-level sketch-based video retrieval (FG-SBVR), where a sketch sequence is used as a query to retrieve a specific target video instance. Compared with sketch-based still image retrieval, and coarse-grained category-level video retrieval, this is more challenging as both visual appearance and motion need to be simultaneously matched at a fine-grained level. We contribute the first FG-SBVR dataset with rich annotations. We then introduce a novel multi-stream multi-modality deep network to perform FG-SBVR under both strong and weakly supervised settings. The key component of the network is a relation module, designed to prevent model over-fitting given scarce training data. We show that this model significantly outperforms a number of existing state-of-the-art models designed for video analysis.
Key for solving fine-grained image categorization is finding discriminate and local regions that correspond to subtle visual traits. Great strides have been made, with complex networks designed specifically to learn part-level discriminate feature representations. In this paper, we show it is possible to cultivate subtle details without the need for overly complicated network designs or training mechanisms -- a single loss is all it takes. The main trick lies with how we delve into individual feature channels early on, as opposed to the convention of starting from a consolidated feature map. The proposed loss function, termed as mutual-channel loss (MC-Loss), consists of two channel-specific components: a discriminality component and a diversity component. The discriminality component forces all feature channels belonging to the same class to be discriminative, through a novel channel-wise attention mechanism. The diversity component additionally constraints channels so that they become mutually exclusive on spatial-wise. The end result is therefore a set of feature channels that each reflects different locally discriminative regions for a specific class. The MC-Loss can be trained end-to-end, without the need for any bounding-box/part annotations, and yields highly discriminative regions during inference. Experimental results show our MC-Loss when implemented on top of common base networks can achieve state-of-the-art performance on all four fine-grained categorization datasets (CUB-Birds, FGVC-Aircraft, Flowers-102, and Stanford-Cars). Ablative studies further demonstrate the superiority of MC-Loss when compared with other recently proposed general-purpose losses for visual classification, on two different base networks. Code available at https://github.com/dongliangchang/Mutual-Channel-Loss
Neural architecture search (NAS) is an emerging topic in machine learning and computer vision. The fundamental ideology of NAS is using an automatic mechanism to replace manual designs for exploring powerful network architectures. One of the key factors of NAS is to scale-up the search space, e.g., increasing the number of operators, so that more possibilities are covered, but existing search algorithms often get lost in a large number of operators. This paper presents a scalable NAS algorithm by designing a factorizable set of architectural parameters, so that the size of the search space goes up quadratically while the burden of optimization increases linearly. As a practical example, we add a set of activation functions to the original set containing convolution, pooling and skip-connect, etc. With a marginal increase in search costs and no extra costs in retraining, we can find interesting architectures that were not explored before and achieve state-of-the-art performance in CIFAR10 and ImageNet, two standard image classification benchmarks.
The item cold-start problem seriously limits the recommendation performance of Collaborative Filtering (CF) methods when new items have either none or very little interactions. To solve this issue, many modern Internet applications propose to predict a new item's interaction from the possessing contents. However, it is difficult to design and learn a map between the item's interaction history and the corresponding contents. In this paper, we apply the Wasserstein distance to address the item cold-start problem. Given item content information, we can calculate the similarity between the interacted items and cold-start ones, so that a user's preference on cold-start items can be inferred by minimizing the Wasserstein distance between the distributions over these two types of items. We further adopt the idea of CF and propose Wasserstein CF (WCF) to improve the recommendation performance on cold-start items. Experimental results demonstrate the superiority of WCF over state-of-the-art approaches.
Collaborative filtering, a widely-used recommendation technique, predicts a user's preference by aggregating the ratings from similar users. As a result, these measures cannot fully utilize the rating information and are not suitable for real world sparse data. To solve these issues, we propose a novel user distance measure named Preference Mover's Distance (PMD) which makes full use of all ratings made by each user. Our proposed PMD can properly measure the distance between a pair of users even if they have no co-rated items. We show that this measure can be cast as an instance of the Earth Mover's Distance, a well-studied transportation problem for which several highly efficient solvers have been developed. Experimental results show that PMD can help achieve superior recommendation accuracy than state-of-the-art methods, especially when training data is very sparse.
Pronouns are often dropped in Chinese sentences, and this happens more frequently in conversational genres as their referents can be easily understood from context. Recovering dropped pronouns is essential to applications such as Information Extraction where the referents of these dropped pronouns need to be resolved, or Machine Translation when Chinese is the source language. In this work, we present a novel end-to-end neural network model to recover dropped pronouns in conversational data. Our model is based on a structured attention mechanism that models the referents of dropped pronouns utilizing both sentence-level and word-level information. Results on three different conversational genres show that our approach achieves a significant improvement over the current state of the art.
Subspace clustering methods based on data self-expression have become very popular for learning from data that lie in a union of low-dimensional linear subspaces. However, the applicability of subspace clustering has been limited because practical visual data in raw form do not necessarily lie in such linear subspaces. On the other hand, while Convolutional Neural Network (ConvNet) has been demonstrated to be a powerful tool for extracting discriminative features from visual data, training such a ConvNet usually requires a large amount of labeled data, which are unavailable in subspace clustering applications. To achieve simultaneous feature learning and subspace clustering, we propose an end-to-end trainable framework, called Self-Supervised Convolutional Subspace Clustering Network (S$^2$ConvSCN), that combines a ConvNet module (for feature learning), a self-expression module (for subspace clustering) and a spectral clustering module (for self-supervision) into a joint optimization framework. Particularly, we introduce a dual self-supervision that exploits the output of spectral clustering to supervise the training of the feature learning module (via a classification loss) and the self-expression module (via a spectral clustering loss). Our experiments on four benchmark datasets show the effectiveness of the dual self-supervision and demonstrate superior performance of our proposed approach.