Narrow Results

We present two neural models for vergence angle control of a robotic head, a simplified and a more complex one. Both models work in a closed-loop manner and do not rely on explicitly computed disparity, but extract the desired vergence angle from the post-processed response of a population of disparity tuned complex cells, the actual gaze direction and the actual vergence angle. The first model assumes that the gaze direction of the robotic head is orthogonal to its baseline and the stimulus is a frontoparallel plane orthogonal to the gaze direction. The second model goes beyond these assumptions, and operates reliably in the general case where all restrictions on the orientation of the gaze, as well as the stimulus position, type and orientation, are dropped.

The ability to recognize the behavior of individuals is of great interest in the general field of safety (e.g. building security, crowd control, transport analysis, independent living for the elderly). Here we report a new real-time acoustic system for human action and behavior recognition that integrates passive audio and active micro-Doppler sonar signatures over multiple time scales. The system architecture is based on a six-layer convolutional neural network, trained and evaluated using a dataset of 10 subjects performing seven different behaviors. Probabilistic combination of system output through time for each modality separately yields 94% (passive audio) and 91% (micro-Doppler sonar) correct behavior classification; probabilistic multimodal integration increases classification performance to 98%. This study supports the efficacy of micro-Doppler sonar systems in characterizing human actions, which can then be efficiently classified using ConvNets. It also demonstrates that the integration of multiple sources of acoustic information can significantly improve the system's performance.

Mathematical morphology has been applied as a collection of nonlinear operations related to object features in images. In this paper, we present morphological layers in deep learning framework, namely MorphNet, to perform atomic morphological operations, such as dilation and erosion. For propagation of losses through the proposed deep learning framework, we approximate the dilation and erosion operations by differential and smooth multivariable functions of the softmax function, and therefore enable the neural network to be optimized. The proposed operations are analyzed by the derivative of approximation functions in the deep learning framework. Experimental results show that the output structuring element of a morphological neuron and the target structuring element are matched to confirm the efficiency and correctness of the proposed framework.

Micro-expression is a kind of brief facial movements which could not be controlled by the nervous system. Micro-expression indicates that a person is hiding his true emotion consciously. Micro-expression recognition has various potential applications in public security and clinical medicine. Researches are focused on the automatic micro-expression recognition, because it is hard to recognize the micro-expression by people themselves. This research proposed a novel algorithm for automatic micro-expression recognition which combined a deep multi-task convolutional network for detecting the facial landmarks and a fused deep convolutional network for estimating the optical flow features of the micro-expression. First, the deep multi-task convolutional network is employed to detect facial landmarks with the manifold-related tasks for dividing the facial region. Furthermore, a fused convolutional network is applied for extracting the optical flow features from the facial regions which contain the muscle changes when the micro-expression appears. Because each video clip has many frames, the original optical flow features of the whole video clip will have high number of dimensions and redundant information. This research revises the optical flow features for reducing the redundant dimensions. Finally, a revised optical flow feature is applied for refining the information of the features and a support vector machine classifier is adopted for recognizing the micro-expression. The main contribution of work is combining the deep multi-task learning neural network and the fusion optical flow network for micro-expression recognition and revising the optical flow features for reducing the redundant dimensions. The results of experiments on two spontaneous micro-expression databases prove that our method achieved competitive performance in micro-expression recognition.

High-throughput chromosome conformation capture (Hi-C) is one of the most popular methods for studying the three-dimensional organization of genomes. However, Hi-C protocols can be expensive since they require large amounts of sample material and may be time-consuming. Most commonly used Hi-C data are low-resolution. Such data can only be used to identify large-scale genomic interactions and are not sufficient to identify the small-scale patterns. We propose a novel deep learning-based computational approach (named ReHiC) that enhances the resolution of Hi-C data and allows us to achieve high-resolution Hi-C data at a relatively low cost. Our model only requires 1/16 down-sampling ratio of the original sequence reading to predict higher resolution Hi-C data. This is very close to high-resolution data in terms of numerical distribution and interaction distribution. More importantly, our framework stacks deeper and converges faster due to residual blocks in the core of the network. Extensive experiments show that ReHiC performs better than HiCPlus and HiCNN, two recently developed and frequently used methods to look at the spatial organization of chromatin structure in the cell. Moreover, the portability of our framework verified by extensive experiments shows that the trained model can also enhance the Hi-C matrix of other cell types efficiently. In conclusion, ReHiC offers more accurate high-resolution image reconstruction in a broad field.

In response to the security management issues of student apartments, a study is conducted on a student apartment access control system based on multitasking cascaded convolutional networks and FaceNet. Firstly, a face detection model is built based on an improved multi-task cascaded convolutional network, and then a face recognition model is built using FaceNet. The results showed that the detection accuracy of the multi-task cascaded convolutional network using the improved non-maximum suppression algorithm was 98.7%, which was higher than the traditional multi-task cascaded convolutional network and effectively improved the detection performance of the multi-task cascaded convolutional network. The face detection model based on the improved multi-task cascaded convolutional network had the shortest average detection time of 361s, the highest average detection accuracy of 90.3%, an accuracy of 99%, a recall rate of 98.5%, and an F1 value of 99%. While maintaining high detection efficiency, it also ensured the accuracy of detection. The average accuracy of the mask detection method based on the MobileNet V2 network was relatively high, at 98.96%. The facial recognition model based on FaceNet achieved a recognition accuracy of 99.15% for faces without masks and 92.04% for faces with masks, with the highest accuracy and recall rates of 99.3% and 99.6%. The model constructed in the study has good application effects in face detection, which helps to improve the security of the student apartment access control system.

Human interaction recognition has been widely studied because it has great scientific importance and many potential practical applications. However, recognizing human interactions is also very challenging especially in realistic environments where the background is dynamic or has varying lighting conditions. Most existing methods rely on either spatio-temporal local features (i.e. SIFT) or human poses, or human joints to model human interactions. As a result, they are not fully unsupervised processes because they require either hand-designed features or human detection results. Motivated by the recent success of deep learning networks, we investigate a three-layer convolutional network which uses the Independent Subspace Analysis (ISA) algorithm to learn hierarchical invariant features from videos. Using the invariant features learned by the ISA, we build a bag-of-features (BOF) model to recognize human interactions. We also evaluate the performance of our approach and the effectiveness of hierarchical invariant features on video sequences of the UT-Interaction dataset which contain both interacting persons and irrelevant pedestrians in the scenes. The dataset imposes several challenging factors including moving backgrounds, clutter scenes, scales and camera jitters. Experimental results show that our three-layer convolutional ISA network is able to learn features which are effective to represent complex activities such as human interactions in realistic environments.