Dr. Santosh Yadav
Multimodal AI Scientist & Entrepreneur

A visionary leader driving cutting-edge advancements in Multimodal AI, Generative AI, Computer Vision, and Conversational AI.

Contact Me
Co-Founder & CTO at BigLook (Cimelium), Ireland
Building Specialised AI for Art Market

Developing a vertical Large Multimodal Model (LMM) and AI agents specifically tailored for the art market, combining technical expertise with commercial vision.

Technical Leadership

Santosh is leading technical strategy and product development, while his co-founder handles commercial development, creating a balanced leadership approach.

1
1
Secured €500K Funding from Enterprise Ireland

This significant funding supports product development and incubation at the ADAPT Centre in Trinity College Dublin, driving our innovative AI solutions forward.

2
2
ConceptionX Deep Tech Venture Partner

Backed by ConceptionX's deeptech venture-building platform, Cimelium pioneers ethical AI for the creative industries—driving innovation where technology meets artistic vision.

Interested in working with us?
  • Currently, we are hiring a Senior Data Scientist. Send your CV and cover letter at santosh.yadav@ieee.org
Major Achievements & Recognition
Commercialisation Fund (June 2025)

Secured €500,000 in funding from Enterprise Ireland to support product development at BigLook, currently incubated at ADAPT Centre in Trinity College Dublin.

Seed Fund Grant (April 2022)

Secured a grant of Rs. 800,000 from the Startup India Seed Fund Scheme to develop an AI-guided personal Yoga assistance platform at Cognix.

Research Technology Innovation (June 2021)

Received $38,000 from LiveSmart Co. Ltd., Tokyo, Japan, to deliver an edge computing system for activity recognition and fall prediction.

TensorFlow Dev Summit 2020 (February 2020)

Selected as one of only 1,000 AI practitioners worldwide to attend this prestigious event at the Google Event Center in Sunnyvale, California.

CSIR - NET/JRF Examination (June 2015)

Secured All India Rank 107 among 300,000 students in Engineering Sciences.

Research Impact
Citation Metrics
1200+
Citations

Total citations of published work

10
H-Index

Impact measurement of scholarly output

10
i10-Index

Publications with 10+ citations

Publication Overview

7 SCI-indexed journal articles in prestigious publications like Neural Networks and Knowledge-Based Systems

7 papers in top international conferences including IEEE WCCI and ICCV Workshop

4 additional manuscripts currently under review in high-impact journals

18+ Total publications and manuscripts, demonstrating a consistent and impactful research output.

Demo Videos

Real-time yoga recognition using deep learning

Skeleton-based human activity recognition using ConvLSTM

Drone-camera based human activity recognition and surveillance

Real-time Karate action recognition and correction feedback using 2 cameras and 3 wearable sensors

YogNet: A two-stream network for real-time multi-person Yoga action recognition and posture correction

ARFDNet: Activity recognition & fall detection system using mobile phone

TBAC: Transformer-based attention consensus for sports activity recognition

Sports video analysis and correction feedback for Gymnastics

Weighted Temporal Attention Module for group activity recognition

Journal Publications
DroneAttention (Neural Networks, Elsevier, 2023)

"DroneAttention: Sparse Weighted Temporal Attention for Drone-Camera Based Activity Recognition"

Authors: Santosh Kumar Yadav, Achleshwar Luthra, Esha Pahwa, Kamlesh Tiwari, Heena Rathore, Hari Mohan Pandey, Peter Corcoran

YogNet (Knowledge-Based Systems, Elsevier, 2022)

"YogNet: A Two-Stream Network for Realtime Yoga Action Recognition and Posture Correction"

Authors: Santosh Kumar Yadav, Aayush Agarwal, Ashish Kumar, Kamlesh Tiwari, Hari Mohan Pandey, Shaik Ali Akbar

CSITime (Neural Networks, Elsevier, 2022)

"CSITime: Privacy-preserving human activity recognition using WiFi channel state information"

Authors: Santosh Kumar Yadav, Siva Sai, Akshay Gundewar, Heena Rathore, Kamlesh Tiwari, Hari Mohan Pandey, Mohit Mathur

ARFDNet (Knowledge-Based Systems, Elsevier, 2022)

"ARFDNet: An Efficient Activity Recognition & Fall Detection System using Latent Feature Pooling"

Authors: Santosh Kumar Yadav, Achleshwar Luthra, Kamlesh Tiwari, Hari Mohan Pandey, Shaik Ali Akbar

Skeleton-based Activity Recognition (Soft Computing, Springer, 2021)

"Skeleton-based human activity recognition using ConvLSTM and guided feature learning"

Authors: Santosh Kumar Yadav, Kamlesh Tiwari, Hari Mohan Pandey, Shaik Ali Akbar

HAR Review Paper (IEEE Transactions, 2020)

"A review of multimodal human activity recognition with special emphasis on classification, applications, challenges and future directions"

Authors: Santosh Kumar Yadav, Kamlesh Tiwari, Hari Mohan Pandey, Shaik Ali Akbar

Realtime Yoga Recognition (Neural Computing and Applications, Springer, 2019)

"Real-time Yoga recognition using deep learning"

Authors: Santosh Kumar Yadav, Amitojdeep Singh, Abhishek Gupta, Jagdish Lal Raheja

Multimodal Activity Recognition (Computer Vision and Image Understanding, Elsevier, 2023)

"A Novel Two Stream Decision Level Fusion of Vision and Inertial Sensors Data for Automatic Multimodal Human Activity Recognition System"

Authors: Santosh Kumar Yadav, Muhtashim Rafiqi, Egna Praneeth Gummana, Kamlesh Tiwari, Hari Mohan Pandey, Shaik Ali Akbar

Yoga Pose Recognition (Computer Vision and Image Understanding, Elsevier, 2024)

"An Efficient Deep Convolutional Neural Network Model for Yoga Pose Recognition using Single Images"

Authors: Santosh Kumar Yadav, Apurv Shukla, Paras Taneja, Kamlesh Tiwari, Hari Mohan Pandey, Shaik Ali Akbar

Light Invariant Action Recognition (IEEE Transaction on Image Understanding, IEEE, 2024)

"LIFT-Net: Dual Stream Fourier Attention for Light Invariant Action Recognition"

Authors: Santosh Kumar Yadav, Onkar Susladkar, Gayatri Deshmukh, Kamlesh Tiwari, Hari Mohan Pandey

Scene Text Understanding (Computer Vision and Image Understanding, Elsevier, 2024)

"CoFormer: Convolutional Fourier Transformer for Scene Text Understanding"

Authors: Onkar Susladkar, Prajwal Gatti, Anand Mishra, Santosh Kumar Yadav

Conference Publications
WTM (IEEE WCCI 2022)

"WTM: Weighted Temporal Attention Module for Group Activity Recognition"

Authors: Santosh Kumar Yadav, Palaash Agarwal, Kamlesh Tiwari, Ehsan Adeli, Hari Mohan Pandey, Shaik Ali Akbar

TBAC (IEEE WCCI 2022)

"TBAC: Transformers Based Attention Consensus for Human Activity Recognition"

Authors: Santosh Kumar Yadav, Shreyas Kera, Raghurama Varma Gonela, Kamlesh Tiwari, Hari Mohan Pandey, Shaik Ali Akbar

MS-KARD (IEEE WCCI 2022)

"MS-KARD: A Benchmark for Multimodal Karate Action Recognition"

Authors: Santosh Kumar Yadav, Aditya Deshmukh, Raghurama Varma Gonela, Shreyas Kera, Kamlesh Tiwari, Hari Mohan Pandey, Shaik Ali Akbar

YogaTube (IEEE WCCI 2022)

"YogaTube: A Video Benchmark for Yoga Action Recognition"

Authors: Santosh Kumar Yadav, Guntaas Singh, Manisha Verma, Kamlesh Tiwari, Hari Mohan Pandey, Shaik Ali Akbar, Peter Corcoran

Eformer (ICCV Workshop 2021)

"Eformer: Edge Enhancement-Based Transformer for Medical Image Denoising"

Authors: Achleshwar Luthra, Harsh Sulakhe, Tanish Mittal, Abhishek Iyer, Santosh Kumar Yadav

SLBERT (IEEE ICASSP 2023)

"SLBERT: A Novel Pre-traning Framework for Joint Speech and Language Modeling"

Authors: Onkar Susladkar, Prajwal Gatti, Santosh Kumar Yadav

Previous Experiences
Senior Data Scientist at WiSAR Labs, ATU Donegal, Ireland

At Data2Sustain (European Digital Innovation Hub) in ATU, Ireland (January 2024 - August 2025), Dr. Yadav designed and implemented scalable AI solutions for startups, handling the complete AI development lifecycle:

Data modeling and pipeline development
API creation and integration
ML model testing and deployment
Technical reporting and documentation
Key Projects Led:
3D Scanning Project

For automated BER assessment

€78,000

Image Processing

Advanced AI-driven solution

€75,000

AI Research Scientist at CeADAR, Dublin, Ireland

During his tenure at CEADAR (Ireland's Centre for Applied AI) in Dublin from March to December 2023, Dr. Yadav developed an automated fruit package detection system for Keeling's using advanced computer vision techniques. His contributions included:

Data Modeling

Created robust data models to handle diverse fruit package types and conditions.

Pipeline Development

Built efficient data processing pipelines for real-time analysis.

API Creation

Developed interfaces for seamless integration with existing systems.

Solution Deployment

Successfully implemented the system in production environments.

Co-Founder & CEO at Cognix, Hyderabad, India

From May 2022 to March 2023, Dr. Yadav co-founded and led Cognix in Hyderabad, India, developing an AI-guided personal assistance platform.

Team Leadership

Managed and inspired a diverse team, fostering a collaborative and high-performance environment.

Strategic Planning

Defined the company's vision and long-term strategy, navigating market trends and competitive landscapes.

Product Development Oversight

Supervised the end-to-end development of the AI-guided personal assistance platform.

Investor Relations & Funding

Successfully engaged with investors and secured critical funding for the startup's growth and operations.

Secured a grant under the Startup India Seed Fund Scheme.

The company was incubated at BITS Pilani.

Rs. 800,000

Postdoctoral Scientist at University of Galway, Ireland

As a Postdoctoral Scientist at the National University of Ireland (NUI) in Galway (August 2021 - March 2023), Dr. Yadav pioneered AI solutions specifically tailored for children's technology.

Low-Cost, Low-Power Design

Developed an AI platform optimized for affordability and energy efficiency.

Multimodal AI Integration

Seamlessly integrated advanced sound and vision capabilities for interactiveness.

Voice-Enabled Toy Applications

Designed the platform for SoapBox Labs, powering sophisticated voice recognition.

AI/ML Internship at Xperi Corporation, Galway, Ireland

During his internship at Xperi Corporation in Galway from August 2021 to March 2023, Dr. Yadav worked on cutting-edge 3D modeling technologies that have applications across gaming, entertainment, and virtual reality industries.

3D Human Body Modeling

Developed advanced techniques for capturing accurate and realistic 3D human body scans.

Synthetic Humans

Created realistic human virtual avatars for various applications in virtual environments.

Project Lead at LiveSmart in Tokyo, Japan

From November 2020 to July 2021, Dr. Yadav served as a project lead and consultant at LiveSmart in Tokyo, Japan, focusing on innovative AI solutions for health monitoring.

Edge Computing for Health Monitoring

As Project Lead and Consultant, Dr. Yadav deployed an edge computing-based, privacy-preserving system for activity recognition and fall prediction. This innovative system helped monitor vulnerable individuals while maintaining their privacy and dignity.

Neurodegenerative Disease Detection

Additionally, he supported the LiveSmart team in developing gait recognition systems to identify walking patterns and balance scoring algorithms using computer vision for early detection of neurodegenerative disease symptoms and progress tracking for physical therapy.

Project Funding

Received to deliver an edge computing system for activity recognition and fall prediction.

$38,000

Research Fellow at CSIR-CEERI, Pilani, India

At CSIR–Central Electronics Engineering Research Institute (CEERI) in Pilani, Rajasthan, India, Dr. Yadav progressed from Junior to Senior Research Fellow, developing expertise in human activity recognition across multiple modalities and applications.

1
Junior Research Fellow (March 2017 - Aug 2019)

Focused on unobtrusive multimodal human activity recognition, including yoga asana recognition and correction, and recognizing activities of daily living.

2
Senior Research Fellow (Aug 2019 - Aug 2021)

Expanded research to diverse applications including karate action recognition, group activity recognition, video transformers, drone cameras, and WiFi-based activity recognition.

Educational Background
1
Ph.D. in Engineering Sciences (Computer Science)

Academy of Scientific and Innovative Research (AcSIR), India
August 2016 - August 2021

Thesis: Unobtrusive Multimodal Human Activity Recognition

2
Bachelor of Technology (B.Tech)

University Institute of Engineering and Technology, Kanpur, India
July 2011 - June 2015

Major: Computer Science and Engineering (CSE)

Connect With Dr. Santosh Yadav
Professional Profiles & Affiliations
  • Personal Website: Explore Dr. Yadav's comprehensive research portfolio, projects, and latest updates on his cutting-edge AI and computer vision work.
  • LinkedIn Profile: Connect with Dr. Yadav for professional networking, industry insights, and updates on his contributions to applied AI.
  • Google Scholar: Access his full list of peer-reviewed academic publications and detailed citation metrics across various high-impact journals and conferences.
  • ResearchGate: Discover his ongoing research activities and preprints, and engage in collaborative projects with the global scientific community.
  • ORCID: A persistent digital identifier ensuring accurate attribution of his extensive research contributions.
  • Scopus: Detailed publication and citation data for his works indexed in the Scopus database.
  • NUI-Galway Profile: Information from his impactful tenure as a postdoctoral scientist at the University of Galway.
  • WiSAR Labs Profile: Insights into his applied research at the WiSAR Applied AI Centre, focusing on industrial applications of AI.
  • CSIR-CEERI Profile: Details on his early research contributions and foundational work in electronics and instrumentation.
  • BITS Pilani: Overview of his doctoral research and contributions during his PhD studies at BITS Pilani.
Contact Information

Email Dr. Yadav directly for inquiries regarding research collaborations, speaking engagements, or professional opportunities.

Emails: santosh.acsir@gmail.com, santosh.yadav@ieee.org

Specialized Research Interests

Generative AI & Agentic AI

Developing next-generation AI systems with advanced generative capabilities and intelligent agents.

Advanced Computer Vision

Creating systems that interpret and understand visual information (image or video) for diverse applications.

Conversational AI

Building natural language interfaces and intelligent agents for human-computer interaction.

Multimodal AI

Integrating data from various sources (e.g., text, image, audio, video) for robust AI systems and applications.

Autonomous Drones

Researching intelligent navigation and surveillance capabilities for UAVs.

Humanoid Robotics

Exploring machine learning algorithms for human-robot interaction and manipulation.

Made with