The 9th National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics (NCVPRIPG)
18-20 July- 2024, IIST Thiruvananthapuram
1 Day Syn-CHBGR Workshop
20 July 2024
18-20 July- 2024, IIST Thiruvananthapuram
20 July 2024
The 2nd workshop Syn-CHBGR-2024 will cover recent developments in contactless hand biometrics and gesture recognition, spanning various practical applications. Recent advancements in generative models within the realms of computer vision and artificial intelligence have revolutionized the way researchers approach data-driven tasks. The advent of sophisticated generative models, such as GANs (Generative Adversarial Networks), VAEs (Variational Autoencoders), or more recently, diffusion models, has empowered practitioners to create synthetic data that closely mirrors real-world scenarios. These models enable the generation of high-fidelity images and sequences, laying the foundation for groundbreaking applications in hand biometrics and gesture analysis. The significance of these generative models lies in their ability to produce synthetic data that is remarkably realistic, thereby mitigating challenges associated with data scarcity and privacy concerns. As a result, the utilization of synthetic data has become increasingly prevalent in biometric domains, offering a versatile and ethical alternative for training and testing the latest deep learning algorithms.
The second edition of the workshop aims to delve into the diverse applications of synthetic data in the realm of hand biometric traits and gesture analysis. Participants will explore how synthetic datasets have been instrumental in training physiological and behavioral biometric recognition systems (fingerprint, finger knuckle image, palmprint, gait, gestures, finger nail image, dorsal palm, finger vein, palm vein), enhancing feature detection and refining recognition algorithms. The workshop will showcase exemplary use cases where the integration of synthetic data has not only overcome data limitations but has also fostered the development of more robust and accurate models. To promote the utilization of synthetic data in biometrics, the second edition of CHBGR is centering its focus on Synthetic Data, aiming to support its adoption and advancement within the field. Topics related to the theme of this workshop continue to draw widespread attention from local, national and international researchers.
Synthetic Data for hand biometrics analysis: hand geometry, finger knuckle, finger nail, fingerprint, finger photo, palm print, palm vein, finger vein, hand expression, signature, keystroke dynamics.
Synthetic data for Multi-biometrics
Synthetic data for Behavioral Biometrics
Synthetic Data for Medical biometrics: EEG, ECG, PPG
Synthetic Data for the Mitigation of Demographic Biases in biometrics
Cross domain biometric recognition: fingerprint Vs finger photo
EMG/ mm-wave Radar hand gesture synthesis and recognition
Hand gesture recognition: cross sensor, cross modality
Demographic effects on physical and behavioral biometric traits
Person identification and re-identification
Presentation attacks and detection in biometrics verification and identification
Morphing attacks and detection
Applications: Forensics, HCI, Healthcare, Image generation
Databases, benchmarking and new protocols, robustness analysis.
Note: Call for Poster/Demo is available here.
Workshop Activities
Expert Talks
Poster presentation (Best poster will be awarded)
GenAI Demo's (Best Demo will be awarded)
Brainstroming session
Speaker from academic and industry
Department of Electrical Engineering,
Indian Institute of Technology Delhi, India
Title: Security and Authentication Issues of Contactless Biometrics
Abstract: Contactless biometric authentication, a form of biometric authentication without physical contact, includes technologies such as iris recognition, facial recognition, and voice recognition. 2D and 3D Contactless fingerprint biometrics has achieved rapid development in the past decades. Biometric authentication serves as the potential solution to addressing identity verification needs, owing to its inherent uniqueness, universality, collectability, and durability. However, we need to have technologies for prevention of misuse and theft of identity, In this talk, I shall address this problem taking the challenges of data constraint scenarios.
Bio: Dr. Santanu Chaudhury is a distinguished academic and researcher in the fields of Computer Vision, Computational Intelligence, Multimedia Systems, and Robotics. He is currently serves as the Schlumberger Chair Professor in the Department of Electrical Engineering at IIT Delhi and was the former Director of IIT Jodhpur. Dr. Chaudhury holds a Ph.D. in Computer Science and Engineering and a B.Tech in Electronics and Electrical Communication Engineering, both from I.I.T Kharagpur. Throughout his illustrious career, Dr. Chaudhury has held several significant positions. He started as a Development Engineer at MMC in the Digital Systems Division, then served as a Junior Scientific Officer at the CSE Department of I.I.T, Delhi. He was also a Computer Engineer at ECSU, ISI Calcutta, and a Lecturer in the Electronics and Electrical Communication Engineering Department at I.I.T, Kharagpur, before becoming an Assistant Professor at I.I.T, Delhi. Additionally, he has chaired the Grades and Registration committee for undergraduate studies at I.I.T, Delhi.
Dr. Chaudhury's research focuses primarily on real-time embedded systems and digital systems, where he has made substantial contributions. His extensive publication record includes numerous research papers in esteemed journals and conferences. He has played a pivotal role in curriculum development at I.I.T, Delhi, including restructuring the B.Tech curriculum introduced in 2003. He also designed and coordinated a certificate course on "Embedded Systems and Applications" for working professionals, which has been successfully run multiple times under the aegis of FITT, I.I.T, Delhi. Furthermore, he is developing a Video Lecture series on "Embedded Systems" supported by ETSC under the NPTEL program. In addition to his academic and research pursuits, Dr. Chaudhury has been actively involved in various extension activities. He has served as a Program Committee Member for IEEE ICCV (2000, 2002, 2005), Program Chair for ICVGIP'98, and Organising Chair for ACCV'2005. He was also a Guest Editor for the IETE Technical Review and served as the General Secretary of IUPRAI (Indian Unit of International Association of Pattern Recognition) from 2005 to 2007. Dr. Chaudhury has been recognized with several prestigious awards and fellowships. He received the INSA Medal for Young Scientists in 1993 and is a Fellow of both the Indian National Academy of Engineers and the National Academy of Sciences. His ongoing contributions continue to shape the field of Electrical Engineering and influence the next generation of engineers and researchers.
Vice President, Samsung R&D, Bangalore
Title: Multimodal Gen AI: VLMs, VLLMs, VLAMs and their applications
Abstract:
Bio: Dr. Balvinder Singh has been serving as Vice President, Research at Samsung Electronics Bangalore since 2009, where he leads Advanced Technology Development and Ecosystem Collaboration for a diverse array of devices, including smartphones, IoT devices, wearables, AR glasses, robotic devices, and automotive systems. His expertise covers a wide range of areas such as Generative AI/ML, the Metaverse (AR/VR/XR), vision, voice, sensor, graph, and text intelligence, multi-modal intelligence, and blockchain technologies. He has played a pivotal role in establishing and building the Center of Excellence (CoE) that is responsible for global leadership in smartphone camera technology.
Prior to his tenure at Samsung, Dr. Singh gained extensive experience at DigiBee Microsystems and Sasken Communication Technologies Ltd. He earned his Ph.D. in Vision, Multimedia Signal, and Image Processing from the Indian Institute of Technology Kanpur, where he studied from 1992 to 1996. He also holds Bachelor's and Master's degrees in Electronics Engineering from the Indian Institute of Technology, Roorkee, completed in 1991. Dr. Singh's diverse skill set includes expertise in Multimodal Vision Language Models, AI Camera & 3D Vision Solutions, and Augmented Reality, Virtual Reality & 360 Camera Products and their applications, making him a valuable contributor to the industry.
Technology Innovation Hub - Human Computer Interaction,
Indian Institute of Technology Mandi
Title: Research and its Translation in the Age of Gen AI
Abstract: Generative AI is evolving from being a mere curiosity to being a core component of value creation. Application of research through its translation can be accelerated with the smart use of Generative AI. This can be achieved if we can focus on addressing the select data challenges and accelerating the transfer knowledge process. This keynote would attempt to link the native strengths associated with Generative AI with the requirements of translational research resulting in the potential acceleration of industry adoption through the quick joining of dots in research. Providing a current view, the keynote will showcase a few compelling examples from the industry thereby illustrating the transformative potential of Generative AI.
Bio: With over 35 years of experience across diverse industries, technologies, and markets, Somjit Amrit has proven himself as a dynamic leader in multicultural global environments, expertly steering organizations as a P&L leader. His career has been marked by pivotal roles in various technology trends, including telecommunications, e-commerce, software services, data science and analytics, the Internet of Things (IoT), and now, Human-Computer Interaction (HCI). As a recognized authority in Business Strategy with a talent for translating technology to meet human needs, Somjit is a frequent speaker at industry conferences. His core expertise lies in applying technology to address real-world problems, a domain where he has consistently delivered impactful solutions.
Somjit excels in fostering a fast-paced, engaging organizational culture, and he is passionate about building, nurturing, and leading high-performance teams. His innovative contributions have been acknowledged with prestigious awards such as the Azim Premji Award for Innovation in 2010 and the Aegis Graham Bell Award for Innovation in IoT in 2018. In his current role as CEO of IIT Mandi iHUB and HCI Foundation, Somjit is committed to establishing a Centre of Excellence in HCI, focusing on translational research, skill development, incubation, and industry collaboration. Under his leadership, the organization is building a high-performance team dedicated to these goals. A voracious reader, Somjit enjoys exploring books, no matter how abstract. In September 2023, he published his first non-fiction book, "Anecdotally Yours," which has garnered critical acclaim. In addition to his executive responsibilities, he also teaches Business Strategy and Technology at premier Institutes of Management. Somjit holds an engineering degree from NIT Rourkela (1984-88) and an MBA from IIM Lucknow (1989-91), qualifications that have laid the foundation for his remarkable career.
Department of Electrical Engineering,
Indian Institute of Technology Indore, India
Title: Effectiveness of Synthetic Data for Fast High-Resolution Fingerprint Recognition
Abstract: High-resolution fingerprint (HRF) recognition is mainly centered around local descriptors created using small image patches around the pores. Although these methods generally provide good verification performance, they are not well-suited for identification due to their poor computational performance and variable and large template size caused by the variable number of useful pore patches in HRF images. This presentation will discuss how the problem mentioned above of variable template size can be overcome by learning to generate fixed-sized global descriptors for the whole HRF image while also taking into account the finer level-3 features by infusing HRF domain knowledge, leading to a deep learning model that is significantly faster than the SOTA. Towards the end, I will also present the results of our study that explored the effectiveness of a synthetic dataset for high-resolution fingerprint recognition.
Bio: Vivek Kanhangad received Ph.D. from the Hong Kong Polytechnic University in 2010. Prior to joining Hong Kong PolyU, he received M.Tech. degree in Electrical Engineering from Indian Institute of Technology Delhi, in 2006 and worked for Motorola India Electronics Ltd, Bangalore for a while. His research interests are in the overlapping areas of computer vision and pattern recognition, focusing on biometrics and biomedical applications. His research team won the Best Iris Liveness Detection with Portable Devices Competition at the International Joint Conference on Biometrics (IJCB 2014), Florida. He regularly serves as a Technical Program Committee Member for reputed international conferences on biometrics. He is currently the Vice President (Education) of the IEEE Biometrics Council, an Editorial Board Member of the IEEE Biometrics Council Newsletter, and a Senior Member of IEEE.
Guidelines (Poster / Demo Presentation)
All posters/ demos should be submitted electronically in PDF format including tables, figures, and references.
The organizers will review each poster/demo submission and inform the acceptance of the poster or demo (with suggestions if any).
At least one author of each accepted poster/demo is required to register for the workshop
For poster presentation, each participant is required to prepare a digital poster A0 size along with a ppt and will be allocated 10 min for presenting their poster followed by 2 min questions/answers.
For the demo, each participant will be allocated 10 min for presenting their demo followed by 2 min questions/answers.
Participant can upload their poster/demo details (only in Pdf format) by 25th June:
Poster/ Demo submission form: Click here for submission
(contact the organizer in case you need any extension/ query)
Associate Professor,
School of Computing and Electrical Engineering,
IIT Mandi
Contact Email: aditya@iitmandi.ac.in
Associate Professor,
School of Computing and Electrical Engineering,
IIT Mandi
Contact Email: arnav@iitmandi.ac.in
Faculty Fellow,
iHub & HCI Foundation,
IIT Mandi
Contact Email: gaurav@tih.iitmandi.ac.in
Indian Institute of Technology Mandi
Dr. Surya Prakash
Indian Institute of Technology Indore
Dr. Kamlash Tiwari
Birla Institute of Technology and Science Pilani
Dr. Ravinder Nath
National Institute of Technology Hamirpur
Poster/ Demo submission: 25 June 2024
Decision to authors: 28 June 2024
Final versions (with changes): 01 July 2024
Full registration for one author for each poster/ demo to be made before: 01 July 2024
Workshop Date - 20 July 2024 (Saturday)
2. Amrutha a, "Detecting Presentation Attacks in Finger Vein Biometrics with Convolutional Autoencoders"
3. Geetanjali et al, "Forehead Creases based User Verification using Spatio-Spatial Temporal Network"
4.
Industry partners for the workshop
For any workshop related query and any academic and industry collaboration or sponsership you can mail at chbgriit@gmail.com