ApproxNet: Content and Contention-Aware Video Object Classification System for Embedded Clients

Videos take a lot of time to transport over the network, hence running analytics on the live video on embedded or mobile devices has become an important system driver. Considering such devices, e.g., surveillance cameras or AR/VR gadgets, are resource constrained, although there has been significant work in creating lightweight deep neural networks (DNNs) for such clients, none of these can adapt to changing runtime conditions, e.g., changes in resource availability on the device, the content characteristics, or requirements from the user. In this article, we introduce ApproxNet, a video object classification system for embedded or mobile clients. It enables novel dynamic approximation techniques to achieve desired inference latency and accuracy trade-off under changing runtime conditions. It achieves this by enabling two approximation knobs within a single DNN model rather than creating and maintaining an ensemble of models, e.g., MCDNN [MobiSys-16]. We show that ApproxNet can adapt seamlessly at runtime to these changes, provides low and stable latency for the image and video frame classification problems, and shows the improvement in accuracy and latency over ResNet [CVPR-16], MCDNN [MobiSys-16], MobileNets [Google-17], NestDNN [MobiCom-18], and MSDNet [ICLR-18].

Visual Feature Learning on Video Object and Human Action Detection: A Systematic Review

Video object and human action detection are applied in many fields, such as video surveillance, face recognition, etc. Video object detection includes object classification and object location within the frame. Human action recognition is the detection of human actions. Usually, video detection is more challenging than image detection, since video frames are often more blurry than images. Moreover, video detection often has other difficulties, such as video defocus, motion blur, part occlusion, etc. Nowadays, the video detection technology is able to implement real-time detection, or high-accurate detection of blurry video frames. In this paper, various video object and human action detection approaches are reviewed and discussed, many of them have performed state-of-the-art results. We mainly review and discuss the classic video detection methods with supervised learning. In addition, the frequently-used video object detection and human action recognition datasets are reviewed. Finally, a summarization of the video detection is represented, e.g., the video object and human action detection methods could be classified into frame-by-frame (frame-based) detection, extracting-key-frame detection and using-temporal-information detection; the methods of utilizing temporal information of adjacent video frames are mainly the optical flow method, Long Short-Term Memory and convolution among adjacent frames.

TASKS AND METHODS OF TEXT SENTIMENT ANALYSIS

The purpose of this article is to study one of the methods of social networks analysis – text sentiment analysis. Today, social media has become a big data base that social network analysis is used for various purposes – from setting up targeted advertising for a cosmetics store to preventing riots at the state level. There are various methods for analyzing social networks such as graph method, text sentiment analysis, audio, and video object analysis. Among them, sentiment analysis is widely used for political, social, consumer research, and also for cybersecurity. Since the analysis of the sentiment of the text involves the analysis of the emotional opinions expressed in the text, the first step is to define the term opinion. An opinion can be simple, that is, a positive, negative or neutral emotion towards a particular object or its aspect. Comparison is also an opinion, but devoid of emotional connotation. To work with simple opinions, the first task of text sentiment analysis is to classify the text. There are three levels of classifications: classification at the text level, at the level of a sentence, and at the aspect level of the object. After classifying the text at the desired level, the next task is to extract structured data from unstructured information. The problem can be solved using the five-tuple method. One of the important elements of a tuple is the aspect in which an opinion is usually expressed. Next, aspect-based sentiment analysis is applied, which involves identifying aspects of the desired object and assessing the polarity of mood for each aspect. This task is divided into two sub-tasks such as aspect extraction and aspect classification. Sentiment analysis has limitations such as the definition of sarcasm and difficulty of working with abbreviated words.