Digital processing of speech signals

We present an algorithm for simultaneous face detection, landmarks localization, pose estimation and gender recognition using deep convolutional neural networks (CNN). The proposed method called, HyperFace, fuses the intermediate layers of a deep CNN using a separate CNN followed by a multi-task learning algorithm that operates on the fused features. It exploits the synergy among the tasks which boosts up their individual performances. Additionally, we propose two variants of HyperFace: (1) HyperFace-ResNet that builds on the ResNet-101 model and achieves significant improvement in performance, and (2) Fast-HyperFace that uses a high recall fast face detector for generating region proposals to improve the speed of the algorithm. Extensive experiments show that the proposed models are able to capture both global and local information in faces and performs significantly better than many competitive algorithms for each of these four tasks.

HyperFace: A Deep Multi-Task Learning Framework for Face Detection, Landmark Localization, Pose Estimation, and Gender Recognition

In recent years, deep learning has garnered tremendous success in a variety of application domains. This new field of machine learning has been growing rapidly and has been applied to most traditional application domains, as well as some new areas that present more opportunities. Different methods have been proposed based on different categories of learning, including supervised, semi-supervised, and un-supervised learning. Experimental results show state-of-the-art performance using deep learning when compared to traditional machine learning approaches in the fields of image processing, computer vision, speech recognition, machine translation, art, medical imaging, medical information processing, robotics and control, bioinformatics, natural language processing, cybersecurity, and many others. This survey presents a brief survey on the advances that have occurred in the area of Deep Learning (DL), starting with the Deep Neural Network (DNN). The survey goes on to cover Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), including Long Short-Term Memory (LSTM) and Gated Recurrent Units (GRU), Auto-Encoder (AE), Deep Belief Network (DBN), Generative Adversarial Network (GAN), and Deep Reinforcement Learning (DRL). Additionally, we have discussed recent developments, such as advanced variant DL techniques based on these DL approaches. This work considers most of the papers published after 2012 from when the history of deep learning began. Furthermore, DL approaches that have been explored and evaluated in different application domains are also included in this survey. We also included recently developed frameworks, SDKs, and benchmark datasets that are used for implementing and evaluating deep learning approaches. There are some surveys that have been published on DL using neural networks and a survey on Reinforcement Learning (RL). However, those papers have not discussed individual advanced techniques for training large-scale deep learning models and the recently developed method of generative models.

A State-of-the-Art Survey on Deep Learning Theory and Architectures

https://hal.archives-ouvertes.fr/hal-01684295/document

Human-aware robot navigation: A survey

Deep learning has demonstrated tremendous success in variety of application domains in the past few years. This new field of machine learning has been growing rapidly and applied in most of the application domains with some new modalities of applications, which helps to open new opportunity. There are different methods have been proposed on different category of learning approaches, which includes supervised, semi-supervised and un-supervised learning. The experimental results show state-of-the-art performance of deep learning over traditional machine learning approaches in the field of Image Processing, Computer Vision, Speech Recognition, Machine Translation, Art, Medical imaging, Medical information processing, Robotics and control, Bio-informatics, Natural Language Processing (NLP), Cyber security, and many more. This report presents a brief survey on development of DL approaches, including Deep Neural Network (DNN), Convolutional Neural Network (CNN), Recurrent Neural Network (RNN) including Long Short Term Memory (LSTM) and Gated Recurrent Units (GRU), Auto-Encoder (AE), Deep Belief Network (DBN), Generative Adversarial Network (GAN), and Deep Reinforcement Learning (DRL). In addition, we have included recent development of proposed advanced variant DL techniques based on the mentioned DL approaches. Furthermore, DL approaches have explored and evaluated in different application domains are also included in this survey. We have also comprised recently developed frameworks, SDKs, and benchmark datasets that are used for implementing and evaluating deep learning approaches. There are some surveys have published on Deep Learning in Neural Networks [1, 38] and a survey on RL [234]. However, those papers have not discussed the individual advanced techniques for training large scale deep learning models and the recently developed method of generative models [1].

The History Began from AlexNet: A Comprehensive Survey on Deep Learning Approaches.

By way of example, the technology disclosed by this document receives image data; extracts a depth image and a color image from the image data; creates a mask image by segmenting the depth image; determines a first likelihood score from the depth image and the mask image using a layered classifier; determines a second likelihood score from the color image and the mask image using a deep convolutional neural network; and determines a class of at least a portion of the image data based on the first likelihood score and the second likelihood score. Further, the technology can pre-filter the mask image using the layered classifier and then use the pre-filtered mask image and the color image to calculate a second likelihood score using the deep convolutional neural network to speed up processing.

Augmenting Layer-Based Object Detection With Deep Convolutional Neural Networks

Advanced driver assistance systems (ADAS) can be significantly improved with effective driver action prediction (DAP). Predicting driver actions early and accurately can help mitigate the effects of potentially unsafe driving behaviors and avoid possible accidents. In this paper, we formulate driver action prediction as a timeseries anomaly prediction problem. While the anomaly (driver actions of interest) detection might be trivial in this context, finding patterns that consistently precede an anomaly requires searching for or extracting features across multi-modal sensory inputs. We present such a driver action prediction system, including a real-time data acquisition, processing and learning framework for predicting future or impending driver action. The proposed system incorporates camera-based knowledge of the driving environment and the driver themselves, in addition to traditional vehicle dynamics. It then uses a deep bidirectional recurrent neural network (DBRNN) to learn the correlation between sensory inputs and impending driver behavior achieving accurate and high horizon action prediction. The proposed system performs better than other existing systems on driver action prediction tasks and can accurately predict key driver actions including acceleration, braking, lane change and turning at durations of 5sec before the action is executed by the driver.

Driver Action Prediction Using Deep (Bidirectional) Recurrent Neural Network.

A method for training a robot to execute a robotic task in a work environment includes moving the robot across its configuration space through multiple states of the task and recording motor schema describing a sequence of behavior of the robot. Sensory data describing performance and state values of the robot is recorded while moving the robot. The method includes detecting perceptual features of objects located in the environment, assigning virtual deictic markers to the detected perceptual features, and using the assigned markers and the recorded motor schema to subsequently control the robot in an automated execution of another robotic task. Markers may be combined to produce a generalized marker. A system includes the robot, a sensor array for detecting the performance and state values, a perceptual sensor for imaging objects in the environment, and an electronic control unit that executes the present method.

Method and system for training a robot using human-assisted task demonstration

We propose a vibrotactile interface in the form of a belt for guiding blind walkers. This interface enables blind walkers to receive haptic directional instructions along complex paths without negatively impacting users’ ability to listen and/or perceive the environment the way some auditory directional instructions do. The belt interface was evaluated in a controlled study with 10 blind individuals and compared to the audio guidance. The experiments were videotaped and the participants’ behaviors and comments were content analyzed. Completion times and deviations from ideal paths were also collected and statistically analyzed. By triangulating the quantitative and qualitative data, we found that the belt resulted in closer path following to the expense of speed. In general, the participants were positive about the use of vibrotactile belt to provide directional guidance.

/pdf/vibrotactile-guidance-for-wayfinding-of-blind-walkers-3iviiz0ugc.pdf

Vibrotactile Guidance for Wayfinding of Blind Walkers

Creating a human-robot interface is a daunting experience. Capabilities and functionalities of the interface are dependent on the robustness of many different sensor and input modalities. For example, object recognition poses problems for state-of-the-art vision systems. Speech recognition in noisy environments remains problematic for acoustic systems. Natural language understanding and dialog are often limited to specific domains and baffled by ambiguous or novel utterances. Plans based on domain-specific tasks limit the applicability of dialog managers. The types of sensors used limit spatial knowledge and understanding, and constrain cognitive issues, such as perspective-taking. In this research, we are integrating several modalities, such as vision, audition, and natural language understanding to leverage the existing strengths of each modality and overcome individual weaknesses. We are using visual, acoustic, and linguistic inputs in various combinations to solve such problems as the disambiguation of referents (objects in the environment), localization of human speakers, and determination of the source of utterances and appropriateness of responses when humans and robots interact. For this research, we limit our consideration to the interaction of two humans and one robot in a retrieval scenario. This paper will describe the system and integration of the various modules prior to future testing.

/pdf/using-vision-acoustics-and-natural-language-for-l3x6jket8h.pdf

Eric Martinson

Papers

Augmenting Layer-Based Object Detection With Deep Convolutional Neural Networks

Driver Action Prediction Using Deep (Bidirectional) Recurrent Neural Network.

Method and system for training a robot using human-assisted task demonstration

Vibrotactile Guidance for Wayfinding of Blind Walkers

Using vision, acoustics, and natural language for disambiguation