Filters








13,415 Hits in 4.5 sec

It's About Time: Analog Clock Reading in the Wild [article]

Charig Yang, Weidi Xie, Andrew Zisserman
2022 arXiv   pre-print
and real data, we leverage the special property of "time", i.e.uniformity, to generate reliable pseudo-labels on real unlabelled clock videos, and show that training on these videos offers further improvements  ...  , we introduce a clock recognition architecture based on spatial transformer networks (STN), which is trained end-to-end for clock alignment and recognition.  ...  Specifically, given an unlabelled clock video, we pass each individual frame through the localisation and recognition module to get the time predictions.  ... 
arXiv:2111.09162v4 fatcat:qjlcjx3azna5heqygtjsb3gud4

Live sports event detection based on broadcast video and web-casting text

Changsheng Xu, Jinjun Wang, Kongwah Wan, Yiqun Li, Lingyu Duan
2006 Proceedings of the 14th annual ACM international conference on Multimedia - MULTIMEDIA '06  
However, the previous approaches are heavily relying on video content itself and require the whole video content for event detection.  ...  We present the framework of our approach and details of text analysis, video analysis and text/video alignment. We conducted experiments on both live games and recorded games.  ...  Game time recognition is to detect the digital clock overlaid on the video and recognize the time from the video clock.  ... 
doi:10.1145/1180639.1180699 dblp:conf/mm/XuWWLD06 fatcat:hyqwcigbrnfzjhgsev6t65747y

Distantly Supervised Semantic Text Detection and Recognition for Broadcast Sports Videos Understanding [article]

Avijit Shah, Topojoy Biswas, Sathish Ramadoss, Deven Santosh Shah
2021 arXiv   pre-print
In this work we study extremely accurate semantic text detection and recognition in sports clocks, and challenges therein.  ...  While both action recognition for multiplayer sports and detection of players has seen robust research, understanding contextual text in video frames still remains one of the most impactful avenues of  ...  (b) Process of end-to-end text recognition from video frames. (c) Effect of scene transitions in contextual object (clock) in contiguous time interval frames.  ... 
arXiv:2111.00629v1 fatcat:qndditc3gvaprj5y47zcamjcoi

Deep Learning-based Automatic Player Identification and Logging in American Football Videos [article]

Hongshan Liu, Colin Aderon, Noah Wagon, Huapu Liu, Steven MacCall, Yu Gan
2022 arXiv   pre-print
Second, we identify players using jersey number recognition with a secondary convolutional neural network, then synchronize it with a game clock subsystem.  ...  We demonstrate the effectiveness and reliability of player identification and the logging system by analyzing the qualitative and quantitative results on football videos.  ...  The subsystem operates on frame-by-frame basis to extract game clock time at the start of a play using the Tesseract for optical character recognition.  ... 
arXiv:2204.13809v1 fatcat:tswjbejxfzdynnahtzik2cuxrm

Reading Both Single and Multiple Digital Video Clocks Using Context-Aware Pixel Periodicity and Deep Learning

Xinguo Yu, Wu Song, Xiaopan Lyu, Bin He, Nan Ye
2020 International Journal of Digital Crime and Forensics  
Reading digital video clocks in real time is a very challenging problem. The first challenge is the clock digit localization.  ...  The first one is that it can read not only single digit video clock but also multiple digit video clocks. The other is that it requires a short length of a video clip.  ...  The clock time plays a critical role in video semantics analysis.  ... 
doi:10.4018/ijdcf.2020040102 fatcat:zcycpqjrpjeatj5t5gxb5lz5iq

An Automated Surveillance System Based on Multi-Processor and GPU Architecture

M. B. Ayed, S. Elkosantini, M. Abid
2017 Engineering, Technology & Applied Science Research  
The aim is to accelerate the system and obtain a reliable and accelerated suspicious behavior recognition.  ...  Video surveillance systems are a powerful tool applied in various systems. Traditional systems based on human vision are to be avoided due to human errors.  ...  The goal is to obtain a reliable detection of suspicious behavior with respect to the real-time constraint for an ATM system.  ... 
doi:10.48084/etasr.1645 fatcat:3fzeawduinastlmfxinfhkceia

Face Recognition Based Attendance System with Auto Alert to Guardian Using Call and SMS

Diksha Ghare, Prajakta Katakdhond, Shraddha Ujgare | Komal Suskar | Prof. Amruta Surana
2018 International Journal of Trend in Scientific Research and Development  
We are implementing attendance system using face recognition. Since face is unique identity of person, the issue of fake attendance and proxies can be solved.  ...  clock-out time.  ...  and clock-out time.  ... 
doi:10.31142/ijtsrd19186 fatcat:xsxqght6efcx3e6yosja3fadau

Page 531 of SMPTE Motion Imaging Journal Vol. 68, Issue 8 [page]

1959 SMPTE Motion Imaging Journal  
Any one video line which originally oc- cupied x ywsec can be decoded in y usec by using a decoding clock frequency x/y times the encoding clock frequency.  ...  If the signal-to- noise ratio is such as to allow reliable recognition of eight discrete amplitude levels, then 1.5 X 106 pulses/sec would carry the digital information.  ... 

Design and Implementation of Vehicle Pedestrian Detection System Based on FPGA

Xue Wen Ma, Shuang Ma, Meng Yao Li, Mei Ling Jin
2011 Advanced Engineering Forum  
The system has better real-time and reliability. At the same time, it has small size, easy to control and the most important is that it has broad application prospects.  ...  By testing, the system can determine a size of 320 × 240 grayscale image takes about 723ms at the clock frequency in the 100MHZ. It achieves the desired functionality.  ...  SDRAM clock must be switched to capture the correct value at the effective time window.  ... 
doi:10.4028/www.scientific.net/aef.2-3.495 fatcat:jav5ct4njjasrcqauwsj73vqje

Capture-to-display delay measurement for visual communication applications

Haoming Chen, Chao Wei, Mingli Song, Ming-Ting Sun, Kevin Lau
2015 APSIPA Transactions on Signal and Information Processing  
The method does not require modifications to the existing system, nor require the encoder and decoder clocks be synchronized.  ...  Furthermore, we propose a solution to solve the multiple-overlapped-timestamp problem due to the exposure time of the camera.  ...  The timestamp with the latest time value is chosen as the time of the current video frame. When no recognition of the QR-code is successful, the system marks its time as "Invalid".  ... 
doi:10.1017/atsip.2015.21 fatcat:glavwvcgerbdvn2xmohqyc7zwm

Study on Data Acquisition and Storage Based on FPGA in Carrier-based Photoelectric Warning System

Feng Qu, Dongjun Yang, Jian Zhao, Qian Sun
2013 Energy and Power Engineering  
In the system, FPGA not only to control the timing of capturing video, also needs to transfers the collected  ...  In order to capture and storage video data real-time for carrier-based photoelectric warning system, an acquisition and storage system based on FPGA is designed.  ...  Computer automatic control technology, image processing and pattern recognition techniques (real-time video enhancement, abnormal target detection and identification automatic target acquisition and tracking  ... 
doi:10.4236/epe.2013.54b110 fatcat:g3r2mbwtvfdevo6byreh53ptta

Page 691 of SMPTE Motion Imaging Journal Vol. 70, Issue 9 [page]

1961 SMPTE Motion Imaging Journal  
Operation by clock time has been allowed for, but in a manner that subordinates the display of clock-time information to a lower priority location on the display panel.  ...  Time Designation The major task of the system is to switch video and audio sources on the air at the proper time.  ... 

Artificial Intelligence Edge Applications in 5G Networks [article]

Carlota Villasante Marcos
2021 arXiv   pre-print
As opposed to earlier architectures, 5G Networks provide low latency access to services with high reliability.  ...  This paper analyzes the feasibility of a real-time Computer Vision use case model in small devices using a fully deployed 5G Network.  ...  OWD(TCP)Uplink = Time (TCP)APP Server -Time (TCP)UE Fig. 4 . 4 Recognition of a chair shown in the image processed by the CV APP Server at 5Tonic.  ... 
arXiv:2109.10267v2 fatcat:zojso4iqrnfdlfilek6g4yrfxu

A Novel Framework for Semantic Annotation and Personalized Retrieval of Sports Video

Changsheng Xu, Jinjun Wang, Hanqing Lu, Yifan Zhang
2008 IEEE transactions on multimedia  
Sports video annotation is important for sports video semantic analysis such as event detection and personalization.  ...  We present the framework of our approach and details of text analysis, video analysis, text/video alignment, and personalized retrieval.  ...  Compared with the traditional methods such as optical character recognition (OCR), our approach is able to achieve the real time performance and the result is more reliable due to the reliable location  ... 
doi:10.1109/tmm.2008.917346 fatcat:w3bbhwkyt5chfbkp4nocemiexe

An Automated Surveillance System based on Multi-Processor System-on-Chip and Hardware Accelerator

Mossaad Ben, SabeurElkosantini, Mohamed Abid
2017 International Journal of Advanced Computer Science and Applications  
The aim is to accelerate the processing and obtain a reliable and accelerated suspicious behavior recognition.  ...  The detection is encountered by complexity and time-consuming process. An abnormal behavior could be identified by different ways: actions, face, trajectory, etc.  ...  The goal is to obtain a reliable detection of suspicious behavior with respect to the real-time constraint for ATM system.  ... 
doi:10.14569/ijacsa.2017.080908 fatcat:2gd7gbshnncgdnrybsbbrhdzia
« Previous Showing results 1 — 15 out of 13,415 results