Vineet Kumar Rakesh

About

Vineet Kumar Rakesh is a doctoral researcher in Computer Science at Homi Bhabha National Institute (HBNI), working at the Variable Energy Cyclotron Centre (VECC), Kolkata, India. His research interests lie in machine learning, computer vision, multimodal artificial intelligence, and ultra-low-bitrate visual communication. His work focuses on talking-head generation, audio-driven facial animation, lip-reading, facial geometry analysis, federated learning, and efficient deep learning systems for real-world communication and vision applications. He is also interested in dataset design, Software Engineering, evaluation protocols, and trustworthy AI practices for generative visual systems. His broader goal is to develop practical, efficient, and scalable AI methods that can operate reliably in resource-constrained environments.

Employment

• Variable Energy Cyclotron Centre Technical Officer
  2003 - Present

Education

• Homi Bhabha National Institute PhD
  2021 - 2026
• West Bengal University of Technology M.Tech.
  2010 - 2012

Projects & Funding

Projects & funding information is unavailable.

Publications (22)

• Analysis of hyperparameter optimization effects on lightweight deep models for real-time image classification
  Scientific Reports 2026 DOI: 10.1038/s41598-026-42748-w
  +Save
• Advancements in talking head generation: a comprehensive review of techniques, metrics, and challenges
  The Visual Computer 2026 DOI: 10.1007/s00371-025-04232-w
  +Save
• BayesFusion-SDF: Probabilistic Signed Distance Fusion with View Planning on CPU
  2026 DOI: 10.48550/ARXIV.2602.19697
  +Save
• Language Independent Phoneme Based Speech Transceiver for Limited Bandwidth Resource Communication
  2026 DOI: 10.1007/978-981-96-9494-5_25
  +Save
• PrivFedTalk: Privacy-Aware Federated Diffusion with Identity-Stable Adapters for Personalized Talking-Head Generation
  2026 DOI: 10.48550/ARXIV.2604.08037
  +Save
• TempoSyncDiff: Distilled Temporally-Consistent Diffusion for Low-Latency Audio-Driven Talking Head Generation
  2026 DOI: 10.48550/ARXIV.2603.06057
  +Save
• VedicTHG: Symbolic Vedic Computation for Low-Resource Talking-Head Generation in Educational Avatars
  2026 DOI: 10.48550/ARXIV.2602.08775
  +Save
• VineetVC: Adaptive Video Conferencing Under Severe Bandwidth Constraints Using Audio-Driven Talking-Head Reconstruction
  2026 DOI: 10.48550/ARXIV.2602.12758
  +Save
• Transparent and Trustworthy Data Practices for Talking Head Generation
  2026 DOI: 10.4018/979-8-2600-1433-2.ch007
  +Save
• Analysis of Hyperparameter Optimization Effects on Lightweight Deep Models for Real-Time Image Classification
  2025 DOI: 10.48550/ARXIV.2507.23315
  +Save