Video action recognition, which is topical in computer vision and video analysis, aims to allocate a short video clip to a pre-defined category such as brushing hair or climbing stairs. Recent works focus on action recognition with deep neural networks that achieve state-of-the-art results in need of high-performance platforms. Despite the fast development of mobile computing, video action recognition on mobile devices has not been fully discussed. In this paper, we focus on the novel mobile video action recognition task, where only the computational capabilities of mobile devices are accessible. Instead of raw videos with huge storage, we choose to extract multiple modalities (including I-frames, motion vectors, and residuals) directly from compressed videos. By employing MobileNetV2 as backbone, we propose a novel Temporal Trilinear Pooling (TTP) module to fuse the multiple modalities for mobile video action recognition. In addition to motion vectors, we also provide a temporal fusion method to explicitly induce the temporal context. The efficiency test on a mobile device indicates that our model can perform mobile video action recognition at about 40FPS. The comparative results on two benchmarks show that our model outperforms existing action recognition methods in model size and time consuming, but with competitive accuracy.
National Key Laboratory for Novel Software Technology（计算机软件新技术国家重点实验室）；Nanjing University（南京大学）
Action recognition is a vital task in computer vision, and many methods are developed to push it to the limit. However, current action recognition models have huge computational costs, which cannot be deployed to real-world tasks on mobile devices. In this paper, we first illustrate the setting of real-time action recognition, which is different from current action recognition inference settings. Under the new inference setting, we investigate state-of-the-art action recognition models on the Kinetics dataset empirically. Our results show that designing efficient real-time action recognition m…
In this paper, we propose a data augmentation method for action recognition using instance segmentation. Although many data augmentation methods have been proposed for image recognition, few of them are tailored for action recognition. Our proposed method, ObjectMix, extracts each object region from two videos using instance segmentation and combines them to create new videos. Experiments on two action recognition datasets, UCF101 and HMDB51, demonstrate the effectiveness of the proposed method and show its superiority over VideoMix, a prior work.
CAS Key Laboratory of Technology in Geo-Spatial Information Processing and Application System（中国科学院地理空间信息处理与应用系统技术重点实验室）；University of Science and Technology of China（中国科学技术大学）
We study the video super-resolution (SR) problem for facilitating video analytics tasks, e.g. action recognition, instead of for visual quality. The popular action recognition methods based on convolutional networks, exemplified by two-stream networks, are not directly applicable on video of low spatial resolution. This can be remedied by performing video SR prior to recognition, which motivates us to improve the SR procedure for recognition accuracy. Tailored for two-stream action recognition networks, we propose two video SR methods for the spatial and temporal streams respectively. On the o…
While many action recognition datasets consist of collections of brief, trimmed videos each containing a relevant action, videos in the real-world (e.g., on YouTube) exhibit very different properties: they are often several minutes long, where brief relevant clips are often interleaved with segments of extended duration containing little change. Applying densely an action recognition system to every temporal clip within such videos is prohibitively expensive. Furthermore, as we show in our experiments, this results in suboptimal recognition accuracy as informative predictions from relevant cli…
In learning action recognition, models are typically pre-trained on object recognition with images, such as ImageNet, and later fine-tuned on target action recognition with videos. This approach has achieved good empirical performance especially with recent transformer-based video architectures. While recently many works aim to design more advanced transformer architectures for action recognition, less effort has been made on how to train video transformers. In this work, we explore several training paradigms and present two findings. First, video transformers benefit from joint training on di…
This notebook paper presents an overview and comparative analysis of our systems designed for the following three tasks in ActivityNet Challenge 2019: trimmed action recognition, dense-captioning events in videos, and spatio-temporal action localization.
Current fully-supervised video datasets consist of only a few hundred thousand videos and fewer than a thousand domain-specific labels. This hinders the progress towards advanced video architectures. This paper presents an in-depth study of using large volumes of web videos for pre-training video models for the task of action recognition. Our primary empirical finding is that pre-training at a very large scale (over 65 million videos), despite on noisy social-media videos and hashtags, substantially improves the state-of-the-art on three challenging public action recognition datasets. Further,…
National University of Singapore（新加坡国立大学）；NEC Corporation（日本电气股份有限公司）
The recent development of commodity 360$^{\circ}$ cameras have enabled a single video to capture an entire scene, which endows promising potentials in surveillance scenarios. However, research in omnidirectional video analysis has lagged behind the hardware advances. In this work, we address the important problem of action recognition in top-view 360$^{\circ}$ videos. Due to the wide filed-of-view, 360$^{\circ}$ videos usually capture multiple people performing actions at the same time. Furthermore, the appearance of people are deformed. The proposed framework first transforms omnidirectional …
Georgia Institute of Technology（佐治亚理工学院）；Carnegie Mellon University（卡内基梅隆大学）
We propose a self-supervised learning method to jointly reason about spatial and temporal context for video recognition. Recent self-supervised approaches have used spatial context [9, 34] as well as temporal coherency [32] but a combination of the two requires extensive preprocessing such as tracking objects through millions of video frames [59] or computing optical flow to determine frame regions with high motion [30]. We propose to combine spatial and temporal context in one self-supervised framework without any heavy preprocessing. We divide multiple video frames into grids of patches and …
National Taiwan University（台湾大学）；Intel Labs, Intel Technologies India Pvt. Ltd.（英特尔印度技术私人有限公司英特尔研究院）；Seoul National University（首尔大学）
We develop a system which generates summaries from seniors' indoor-activity videos captured by a social robot to help remote family members know their seniors' daily activities at home. Unlike the traditional video summarization datasets, indoor videos captured from a moving robot poses additional challenges, namely, (i) the video sequences are very long (ii) a significant number of video-frames contain no-subject or with subjects at ill-posed locations and scales (iii) most of the well-posed frames contain highly redundant information. To address this problem, we propose to \hl{exploit} pose …
Vision-based activity recognition is essential for security, monitoring and surveillance applications. Further, real-time analysis having low-quality video and contain less information about surrounding due to poor illumination, and occlusions. Therefore, it needs a more robust and integrated model for low quality and night security operations. In this context, we proposed a hybrid model for illumination invariant human activity recognition based on sub-image histogram equalization enhancement and k-key pose human silhouettes. This feature vector gives good average recognition accuracy on thre…
Fondazione Bruno Kessler（布鲁诺·凯斯勒基金会）；University of Trento（特伦托大学）
Most recent approaches for action recognition from video leverage deep architectures to encode the video clip into a fixed length representation vector that is then used for classification. For this to be successful, the network must be capable of suppressing irrelevant scene background and extract the representation from the most discriminative part of the video. Our contribution builds on the observation that spatio-temporal patterns characterizing actions in videos are highly correlated with objects and their location in the video. We propose Top-down Attention Action VLAD (TA-VLAD), a deep…
This paper presents a review of human activity recognition and behaviour understanding in video sequence. The key objective of this paper is to provide a general review on the overall process of a surveillance system used in the current trend. Visual surveillance system is directed on automatic identification of events of interest, especially on tracking and classification of moving objects. The processing step of the video surveillance system includes the following stages: Surrounding model, object representation, object tracking, activity recognition and behaviour understanding. It describes…
Static appearance of video may impede the ability of a deep neural network to learn motion-relevant features in video action recognition. In this paper, we introduce a new concept, Dynamic Appearance (DA), summarizing the appearance information relating to movement in a video while filtering out the static information considered unrelated to motion. We consider distilling the dynamic appearance from raw video data as a means of efficient video understanding. To this end, we propose the Pixel-Wise Temporal Projection (PWTP), which projects the static appearance of a video into a subspace within…
Video action recognition is one of the representative tasks for video understanding. Over the last decade, we have witnessed great advancements in video action recognition thanks to the emergence of deep learning. But we also encountered new challenges, including modeling long-range temporal information in videos, high computation costs, and incomparable results due to datasets and evaluation protocol variances. In this paper, we provide a comprehensive survey of over 200 existing papers on deep learning for video action recognition. We first introduce the 17 video action recognition datasets …
Kapital Bank（资本银行）；Khazar University（哈扎尔大学）；Consumer Experience Tribe（消费者体验部落）；Lumos Extension（Lumos扩展）；School of Data（数据学院）
Recognizing human actions based on videos has became one of the most popular areas of research in computer vision in recent years. This area has many applications such as surveillance, robotics, health care, video search and human-computer interaction. There are many problems associated with recognizing human actions in videos such as cluttered backgrounds, obstructions, viewpoints variation, execution speed and camera movement. A large number of methods have been proposed to solve the problems. This paper focus on spatial and temporal pattern recognition for the classification of videos using…
Efficient video action recognition remains a challenging problem. One large model after another takes the place of the state-of-the-art on the Kinetics dataset, but real-world efficiency evaluations are often lacking. In this work, we fill this gap and investigate the use of transformers for efficient action recognition. We propose a novel, lightweight action recognition architecture, VideoLightFormer. In a factorized fashion, we carefully extend the 2D convolutional Temporal Segment Network with transformers, while maintaining spatial and temporal video structure throughout the entire model. …
Information Technology University（信息技术大学）；University of North Carolina at Charlotte（北卡罗来纳大学夏洛特分校）
The goal of human action recognition is to temporally or spatially localize the human action of interest in video sequences. Temporal localization (i.e. indicating the start and end frames of the action in a video) is referred to as frame-level detection. Spatial localization, which is more challenging, means to identify the pixels within each action frame that correspond to the action. This setting is usually referred to as pixel-level detection. In this chapter, we are using action, activity, event interchangeably.
We present a new action recognition deep neural network which adaptively learns the best action velocities in addition to the classification. While deep neural networks have reached maturity for image understanding tasks, we are still exploring network topologies and features to handle the richer environment of video clips. Here, we tackle the problem of multiple velocities in action recognition, and provide state-of-the-art results for gesture recognition, on known and new collected datasets. We further provide the training steps for our semi-supervised network, suited to learn from huge unla…
In this paper, we examined the zero-shot activity recognition task with the usage of videos. We introduce an auto-encoder based model to construct a multimodal joint embedding space between the visual and textual manifolds. On the visual side, we used activity videos and a state-of-the-art 3D convolutional action recognition network to extract the features. On the textual side, we worked with GloVe word embeddings. The zero-shot recognition results are evaluated by top-n accuracy. Then, the manifold learning ability is measured by mean Nearest Neighbor Overlap. In the end, we provide an extens…
Transformer-based models have achieved top performance on major video recognition benchmarks. Benefiting from the self-attention mechanism, these models show stronger ability of modeling long-range dependencies compared to CNN-based models. However, significant computation overheads, resulted from the quadratic complexity of self-attention on top of a tremendous number of tokens, limit the use of existing video transformers in applications with limited resources like mobile devices. In this paper, we extend Mobile-Former to Video Mobile-Former, which decouples the video architecture into a lig…
MobileVidFactory: Automatic Diffusion-Based Social Media Video Generation for Mobile Devices from Text
University of Electronic Science and Technology of China（电子科技大学）；Microsoft Research（微软研究院）；University of Rochester（罗切斯特大学）；National Taiwan University（国立台湾大学）
Northwestern Polytechnical University（西北工业大学）；City University of Hong Kong（香港城市大学）
The ubiquity of camera-embedded devices and the advances in deep learning have stimulated various intelligent mobile video applications. These applications often demand on-device processing of video streams to deliver real-time, high-quality services for privacy and robustness concerns. However, the performance of these applications is constrained by the raw video streams, which tend to be taken with small-aperture cameras of ubiquitous mobile platforms in dim light. Despite extensive low-light video enhancement solutions, they are unfit for deployment to mobile devices due to their complex mo…
Istanbul Technical University（伊斯坦布尔理工大学）；Hamad bin Khalifa University（哈马德·本·哈利法大学）
Food recognition is an important task for a variety of applications, including managing health conditions and assisting visually impaired people. Several food recognition studies have focused on generic types of food or specific cuisines, however, food recognition with respect to Middle Eastern cuisines has remained unexplored. Therefore, in this paper we focus on developing a mobile friendly, Middle Eastern cuisine focused food recognition application for assisted living purposes. In order to enable a low-latency, high-accuracy food classification system, we opted to utilize the Mobilenet-v2 …
Research and Academic Computer Network（研究与学术计算机网络）；Warsaw University of Technology（华沙理工大学）
With the recent shift towards mobile computing, new challenges for biometric authentication appear on the horizon. This paper provides a comprehensive study of cross-spectral iris recognition in a scenario, in which high quality color images obtained with a mobile phone are used against enrollment images collected in typical, near-infrared setups. Grayscale conversion of the color images that employs selective RGB channel choice depending on the iris coloration is shown to improve the recognition accuracy for some combinations of eye colors and matching software, when compared to using the red…
Concordia University（康考迪亚大学）；University of Arkansas（阿肯色大学）；Carnegie Mellon University（卡内基梅隆大学）
Deep neural networks have been widely used in numerous computer vision applications, particularly in face recognition. However, deploying deep neural network face recognition on mobile devices has recently become a trend but still limited since most high-accuracy deep models are both time and GPU consumption in the inference stage. Therefore, developing a lightweight deep neural network is one of the most practical solutions to deploy face recognition on mobile devices. Such the lightweight deep neural network requires efficient memory with small number of weights representation and low cost o…
Mobile virtual reality (VR) head mounted displays (HMD) have become popular among consumers in recent years. In this work, we demonstrate real-time egocentric hand gesture detection and localization on mobile HMDs. Our main contributions are: 1) A novel mixed-reality data collection tool to automatic annotate bounding boxes and gesture labels; 2) The largest-to-date egocentric hand gesture and bounding box dataset with more than 400,000 annotated frames; 3) A neural network that runs real time on modern mobile CPUs, and achieves higher than 76% precision on gesture recognition across 8 classes…
The target of this research is to experiment, iterate and recommend a system that is successful in recognition of American Sign Language (ASL). It is a challenging as well as an interesting problem that if solved will bring a leap in social and technological aspects alike. In this paper, we propose a real-time recognizer of ASL based on a mobile platform, so that it will have more accessibility and provides an ease of use. The technique implemented is Transfer Learning of new data of Hand gestures for alphabets in ASL to be modelled on various pre-trained high- end models and optimize the best…
Despite the recent success of video object detection on Desktop GPUs, its architecture is still far too heavy for mobiles. It is also unclear whether the key principles of sparse feature propagation and multi-frame feature aggregation apply at very limited computational resources. In this paper, we present a light weight network architecture for video object detection on mobiles. Light weight image object detector is applied on sparse key frames. A very small network, Light Flow, is designed for establishing correspondence across frames. A flow-guided GRU module is designed to effectively aggr…
Mobile biometric approaches provide the convenience of secure authentication with an omnipresent technology. However, this brings an additional challenge of recognizing biometric patterns in unconstrained environment including variations in mobile camera sensors, illumination conditions, and capture distance. To address the heterogeneous challenge, this research presents a novel heterogeneity aware loss function within a deep learning framework. The effectiveness of the proposed loss function is evaluated for periocular biometrics using the CSIP, IMP and VISOB mobile periocular databases. The …
This paper introduces an online model for object detection in videos designed to run in real-time on low-powered mobile and embedded devices. Our approach combines fast single-image object detection with convolutional long short term memory (LSTM) layers to create an interweaved recurrent-convolutional architecture. Additionally, we propose an efficient Bottleneck-LSTM layer that significantly reduces computational cost compared to regular LSTMs. Our network achieves temporal awareness by using Bottleneck-LSTMs to refine and propagate feature maps across frames. This approach is substantially …
Dayananda Sagar College of Engineering（达亚南达·萨加尔工程学院）
Object recognition in the video sequence or images is one of the sub-field of computer vision. Moving object recognition from a video sequence is an appealing topic with applications in various areas such as airport safety, intrusion surveillance, video monitoring, intelligent highway, etc. Moving object recognition is the most challenging task in intelligent video surveillance system. In this regard, many techniques have been proposed based on different methods. Despite of its importance, moving object recognition in complex environments is still far from being completely solved for low resol…
Imperial College London（帝国理工学院）；Stony Brook University, SUNY（纽约州立大学石溪分校）
We address the task of recognizing objects from video input. This important problem is relatively unexplored, compared with image-based object recognition. To this end, we make the following contributions. First, we introduce two comprehensive datasets for video-based object recognition. Second, we propose Latent Bi-constraint SVM (LBSVM), a maximum-margin framework for video-based object recognition. LBSVM is based on Structured-Output SVM, but extends it to handle noisy video data and ensure consistency of the output decision throughout time. We apply LBSVM to recognize office objects and mu…
Fudan University（复旦大学）；Inception Institute of Artificial Intelligence（起源人工智能研究院）；University of Maryland（马里兰大学）
Label distributions in real-world are oftentimes long-tailed and imbalanced, resulting in biased models towards dominant labels. While long-tailed recognition has been extensively studied for image classification tasks, limited effort has been made for video domain. In this paper, we introduce VideoLT, a large-scale long-tailed video recognition dataset, as a step toward real-world video recognition. Our VideoLT contains 256,218 untrimmed videos, annotated into 1,004 classes with a long-tailed distribution. Through extensive studies, we demonstrate that state-of-the-art methods used for long-t…
Uncertainty-aware Bridge based Mobile-Former Network for Event-based Pattern Recognition
School of Computer Science and Technology, Anhui University（安徽大学计算机科学与技术学院）；School of Electronic and Information Engineering, Anhui University（安徽大学电子与信息工程学院）；Suzhou Glink IoT Technology Co., Ltd.（苏州格联物联网科技有限公司）
In this paper, we introduce a challenging new dataset, MLB-YouTube, designed for fine-grained activity detection. The dataset contains two settings: segmented video classification as well as activity detection in continuous videos. We experimentally compare various recognition approaches capturing temporal structure in activity videos, by classifying segmented videos and extending those approaches to continuous videos. We also compare models on the extremely difficult task of predicting pitch speed and pitch type from broadcast baseball videos. We find that learning temporal structure is valua…
Although deep learning approaches have achieved performance surpassing humans for still image-based face recognition, unconstrained video-based face recognition is still a challenging task due to large volume of data to be processed and intra/inter-video variations on pose, illumination, occlusion, scene, blur, video quality, etc. In this work, we consider challenging scenarios for unconstrained video-based face recognition from multiple-shot videos and surveillance videos with low-quality frames. To handle these problems, we propose a robust and efficient system for unconstrained video-based …
This work discusses and implements the application of speaker recognition for the detection of collaborations in YouTube videos. CATANA, an existing framework for detection and analysis of YouTube collaborations, is utilizing face recognition for the detection of collaborators, which naturally performs poor on video-content without appearing faces. This work proposes an extension of CATANA using active speaker detection and speaker recognition to improve the detection accuracy.
Identity of a vehicle is done through the vehicle license plate by traffic police in general. Au- tomatic vehicle license plate recognition has several applications in intelligent traffic management systems. The security situation across the globe and particularly in India demands a need to equip the traffic police with a system that enables them to get instant details of a vehicle. The system should be easy to use, should be mobile, and work 24 x 7. In this paper, we describe a mobile phone based, client-server architected, license plate recognition system. While we use the state of the art i…
The growth of high-performance mobile devices has resulted in more research into on-device image recognition. The research problems are the latency and accuracy of automatic recognition, which remains obstacles to its real-world usage. Although the recently developed deep neural networks can achieve accuracy comparable to that of a human user, some of them still lack the necessary latency. This paper describes the development of the architecture of a new convolutional neural network model, NU-LiteNet. For this, SqueezeNet was developed to reduce the model size to a degree suitable for smartpho…
NAVER Clova；KAIST（韩国科学技术院）；NAVER AI Lab（NAVER人工智能实验室）；Yonsei University（延世大学）
Data augmentation has recently emerged as an essential component of modern training recipes for visual recognition tasks. However, data augmentation for video recognition has been rarely explored despite its effectiveness. Few existing augmentation recipes for video recognition naively extend the image augmentation methods by applying the same operations to the whole video frames. Our main idea is that the magnitude of augmentation operations for each frame needs to be changed over time to capture the real-world video's temporal variations. These variations should be generated as diverse as po…
Beihang University（北京航空航天大学）；China University of Mining and Technology（中国矿业大学）；University of Central Florida（中佛罗里达大学）；Nortumbria Univesity（诺森比亚大学）
Gesture recognition is a challenging problem in the field of biometrics. In this paper, we integrate Fisher criterion into Bidirectional Long-Short Term Memory (BLSTM) network and Bidirectional Gated Recurrent Unit (BGRU),thus leading to two new deep models termed as F-BLSTM and F-BGRU. BothFisher discriminative deep models can effectively classify the gesture based on analyzing the acceleration and angular velocity data of the human gestures. Moreover, we collect a large Mobile Gesture Database (MGD) based on the accelerations and angular velocities containing 5547 sequences of 12 gestures. E…
University of Science and Technology of China（中国科学技术大学）；Microsoft Research（微软研究院）；Microsoft Research Asia（微软亚洲研究院）
Deep convolutional neutral networks have achieved great success on image recognition tasks. Yet, it is non-trivial to transfer the state-of-the-art image recognition networks to videos as per-frame evaluation is too slow and unaffordable. We present deep feature flow, a fast and accurate framework for video recognition. It runs the expensive convolutional sub-network only on sparse key frames and propagates their deep feature maps to other frames via a flow field. It achieves significant speedup as flow computation is relatively fast. The end-to-end training of the whole architecture significa…
Halmstad University（哈尔姆斯塔德大学）；University of Balearic Islands（巴利阿里群岛大学）
We address the use of selfie ocular images captured with smartphones to estimate age and gender. Partial face occlusion has become an issue due to the mandatory use of face masks. Also, the use of mobile devices has exploded, with the pandemic further accelerating the migration to digital services. However, state-of-the-art solutions in related tasks such as identity or expression recognition employ large Convolutional Neural Networks, whose use in mobile devices is infeasible due to hardware limitations and size restrictions of downloadable applications. To counteract this, we adapt two exist…
Hochschule Darmstadt（达姆施塔特应用科学大学）；Freie Universität Berlin（柏林自由大学）
This work presents an automated touchless fingerprint recognition system for smartphones. We provide a comprehensive description of the entire recognition pipeline and discuss important requirements for a fully automated capturing system. Also, our implementation is made publicly available for research purposes. During a database acquisition, a total number of 1,360 touchless and touch-based samples of 29 subjects are captured in two different environmental situations. Experiments on the acquired database show a comparable performance of our touchless scheme and the touch-based baseline scheme…
We present SlowFast networks for video recognition. Our model involves (i) a Slow pathway, operating at low frame rate, to capture spatial semantics, and (ii) a Fast pathway, operating at high frame rate, to capture motion at fine temporal resolution. The Fast pathway can be made very lightweight by reducing its channel capacity, yet can learn useful temporal information for video recognition. Our models achieve strong performance for both action classification and detection in video, and large improvements are pin-pointed as contributions by our SlowFast concept. We report state-of-the-art ac…
School of Computing（计算机学院）；Wichita State University（威奇托州立大学）
A number of studies have demonstrated the efficacy of deep learning convolutional neural network (CNN) models for ocular-based user recognition in mobile devices. However, these high-performing networks have enormous space and computational complexity due to the millions of parameters and computations involved. These requirements make the deployment of deep learning models to resource-constrained mobile devices challenging. To this end, only a handful of studies based on knowledge distillation and patch-based models have been proposed to obtain compact size CNN models for ocular recognition in…
New York University（纽约大学）；Wikimedia Foundation（维基媒体基金会）；University of Turin（都灵大学）
Rendering Wikipedia content through mobile and augmented reality mediums can enable new forms of interaction in urban-focused user communities facilitating learning, communication and knowledge exchange. With this objective in mind, in this work we develop a mobile application that allows for the recognition of notable sites featured on Wikipedia. The application is powered by a deep neural network that has been trained on crowd-sourced imagery describing sites of interest, such as buildings, statues, museums or other physical entities that are present and visually accessible in an urban envir…
