Hojin Cho, an algorithm engineer of StradVision, will serve as a member of the Program Committee of the Fourteenth IAPR International Conference on Document Analysis and Recognition (ICDAR 2017) conference.
StradVision will give a talk titled "SVNet, a CNN-based object detection for ADAS" at NVIDIA GTC 2017 in San Jose, CA. Also, StradVision will demo SVNet on car by appointment. Please contact firstname.lastname@example.org if you are interested in seeing the car demo.
StradVision will showcase their object detection SW on various embedded HW at the Embedded Vision Summit 2017 in Santa Clara, CA. Also, StradVision will demo SVNet on car by appointment. Please contact email@example.com if you are interested in seeing the car demo.
StradVision showcased their object detection SW on Qualcomm's S820A at Thundersoft's booth at CES 2017 in Las Vegas, NV. Also, StradVision demoed SVNet on car by appointment.
StradVision will give a talk titled "CNN-based Object Detection and its acceleration on a vision DSP for ADAS and Surveillance" at CEVA TECHNOLOGY SYMPOSIUM 2016 – ASIA (CEVA 2016 技术研讨会 - 亚洲) in Hsinchu, Shenzhen, and Shanghai on 21, 23, and 25 Nov respectively. Read More (In Chinese or English).
StradVision will present Object Detection & Recognition Software for ADAS on R-Car V2H at R-Car Forum 2016 in Tokyo, Japan.
StradVision will present Object Detection & Recognition Software for ADAS and Scene Text Detection & Recognition "anywhere anytime translation instantly" at 5C97 in CPS/IoT Technology & Software area at CEATEC 2016 in Chiba, Japan. Read More (In Japanese or English). Find the poster (In Japanese).
SYSTRAN International presented their translation app with StradVision’s text detection & recognition software at N2.C30 at MWC Shanghai 2016. Read More.
StradVision joined R-Car Consortium of Renesas as a Middleware/Applications partner. See More.
StradVision will showcase their pedestrian detection on TI TDA2x board at the Embedded Vision Summit 2016 in Santa Clara, CA. Read More.
Taewoong Jang, an algorithm engineer at StradVision, will present his work with his colleagues outside StradVision, on transfer learning with deep representation at CVPR 2016 (Oral), one of the major conferences in computer vision.
Hojin Cho, Myungchul Sung, and Bongjin Jun will present their work on scene text detection algorithm at CVPR 2016, one of the major conferences in computer vision.
StradVision presented Autonomous Vision Detection with KT in GSMA Innovation City at Hall 3 Stand 3A11 at MWC 2016. Watch More.
StradVision joined Renesas Korea Alliance Partner as a software partner.
StradVision will demo and pitch their ADAS software at NAVER D2 Startup Factory. Read More (in Korean).
StradVision pitched their automotive solution at KPMG's 11th Annual Automotive Executive Share Forum Series at the 2016 North American International Auto Show in Detroit, MI.
StradVision's patent application was issued as U.S. Patent No. 9,230,193 on 5 Jan 2016 and is entitled "Method for increasing object detection rate or object recognition rate and classifier using locally decorrelated channel feature (LDCF)". Please find the patent.
StradVision pitched at the 2nd GiGA IoT Alliance Makers Day. Read More (in Korean).
StradVision pitched their automotive solution at Autotech Council meeting in San Jose, CA.
StradVision pitched and showcased their pedestrian detection on TI TDA2x board at the Embedded Vision Alliance meeting in San Jose, CA.
StradVision pitched and showcased their automotive solution at the Demo Day of Fall 2015 season of Orange Fab Asia in Tokyo.
StradVision, a company started by former Samsung SDS and Intel Korea engineer Jun Bong Jin, who is trying to crack one of computing’s hardest goals: how to get robots to recognize things by sight.
Cars that stop automatically for pedestrians and smart glass that translates languages are now closer to reality than fantasy.
StradVision's patent application was issued as U.S. Patent No. 9,141,883 on 22 Sep 2015 and is entitled "Method, hard negative proposer, and classifier for supporting to collect hard negative images using a similarity map". Please find the patent.
Results of the ICDAR 2015 Robust Reading Competition are presented. StradVision achieved the first ranks in 4 out of the 7 newly added tasks, i.e.,
Task 1.4 (Born Digital End-to-End, Strong), Task 2.4 (Focused Text End-to-End, Weak),
Task 4.1 (Incidental Text Localisation), and Task 4.4 (Incidental Text End-to-End, Strong).
Please find the technical report by the organizer.
StradVision joined the ICDAR 2015 Competition on Text Reading in the Wild.
There were several teams registered for the competition, but StradVision was the only one that submitted valid results for the detection task.
Please find the sample results and the technical report by the organizer.
CogniVue offers the most comprehensive embedded vision solution in the market today. StradVision joined Cognivue’s APEX Partner Program.
Myunchul Sung, Bongjin Jun, Hojin Cho, Prof. Daijin Kim, all from StradVision, will present
(ORAL) their latest work on scene text detection algorithm at International Conference on Document Analysis and Recognition (ICDAR) 2015.
The paper is available here.
The Embedded Vision Alliance is a collaboration to enable rapid growth of practical computer vision features in a diversity of systems and associated software.
StradVision became a Technical Member of the Alliance on May 2015
ARM is the world's leading semiconductor IP company. StradVision became a member of ARM Connected Community program on Apr 2015.
StradVision submitted its scene text detection and recognition results for Challenge 1 thru 4. Results will be announced during the ICDAR conference this summer.
We have a preview results for Task 1 (Rank 1) and Task 2 (Rank 3) in Challenge 1,
and Task 1 (Rank 1) and Task 2 (Rank 1) in Challenge 2
StradVision will present its latest advances in embedded computer vision in MWC – Hall 8.0 Stand 8.0E36.
Our product and technology portfolio includes embedded vision software such as scene text detection & recognition, pedestrian detection,
head & shoulder detection implemented in various smart glasses and IOT devices.
Woonhyun Nam, an algorithm engineer of StradVision, presented his latest work on pedestrian detection algorithm at NIPS 2014,
one of the top conferences in machine learning. For more details, please find his paper