Vivek Anand

Awarded the prestigious Fellowship for Academic and Research Excellence (FARE) for outstanding PhD research contributions, recognizing exceptional academic achievement.

Developed ReaLiTy, a scalable physics-informed sim-to-real pipeline, transforming simulated LiDAR data into realistic, sensor and weather-specific representations for autonomous driving simulators.

Released LADS, a comprehensive suite of physically grounded benchmark LiDAR datasets, significantly advancing reproducible research in LiDAR realism and adverse weather degradation.

Investigated autoencoder-based diffusion models for simulation-to-reality transfer in RGB imagery, improving realism and domain alignment for autonomous driving applications.

Leveraged geometry-aware generative techniques and disentangled representation learning to enhance visual fidelity and controllability in simulated visual data.

Designed complex PhD-level research and reasoning problems to rigorously evaluate the capabilities of advanced AI models.

Provided structured, expert-level feedback to significantly improve AI model reasoning quality and alignment with sophisticated analytical expectations.

Contributed to the strategic development of AI evaluation methodologies, enhancing the robustness and reliability of cutting-edge AI systems.

Led the institute-wide PhD placement drive, successfully coordinating departmental teams and engaging over 20 industry partners to secure career opportunities for PhD candidates.

Streamlined communication channels between students, faculty, and recruiters, improving placement efficiency by 15% through strategic coordination.

Organized career workshops and networking events, preparing over 100 PhD students for successful transitions into professional roles.

Developed a novel physics-informed cycle-consistent learning architecture for realistic LiDAR intensity simulation, enabling sensor-aware modeling across real and simulated data without paired supervision.

Extended the framework to effectively model adverse weather conditions by coupling physics-based scattering models with learning-based approaches for realistic snow and rain intensity degradation.

Published multiple peer-reviewed papers in top-tier journals and conferences, disseminating cutting-edge research findings in LiDAR simulation and AI.

Delivered impactful guest lectures at the NGP-DST Winter School on Geospatial Technologies, effectively disseminating specialized knowledge to a diverse audience of researchers and students.

Engaged participants in discussions on cutting-edge geospatial techniques, fostering deeper understanding and interest in the field.

Contributed to the curriculum development for the winter school, ensuring the inclusion of relevant and timely topics in geospatial science.

Provided comprehensive teaching assistance for Geoinformatics and Digital Image Processing, including support for adjustment computation, enhancing student comprehension and practical skills.

Facilitated laboratory sessions and graded assignments, contributing to the academic success of over 50 students per semester.

Developed supplementary learning materials and conducted review sessions, improving student performance in complex technical subjects.

Adapted existing CNN and GAN-based architectures into physics-aware generative models, enabling advanced data-driven simulation of LiDAR data.

Incorporated key physical priors, including range, incidence angle, and material reflectance, directly into generative learning pipelines, enhancing model accuracy and realism.

Collaborated with international researchers to advance the state-of-the-art in physics-informed AI for geospatial applications.

Applied advanced geospatial analysis techniques to crowdsourced Cycle-to-Work data, successfully identifying critical demand patterns for urban cycling infrastructure development.

Generated data-driven insights that informed strategic urban planning initiatives, contributing to sustainable city development.

Collaborated with a multidisciplinary team to translate complex spatial data into actionable recommendations for urban mobility projects.