Nirav Savaliya

Nirav Savaliya

Sr. Robotics Research Engineer

Honda Research Institute

Research Interests

Multi-Modal Spatial Perception
3D Scene Understanding
Autonomous Driving / BEV Perception
Mapping & Localization / SLAM

About

I'm a Senior Research Engineer at Honda Research Institute in San Jose, CA.

My current research focus is on Spatial Perception for Embodied AI Agents, specifically focusing on Visual, natural language, auditory, and motion modalities.

I also have experience working on SLAM and Bird Eye View Perception for Autonomous Vehicles.

Selected Publications

View All →

FAST-EQA: Efficient Embodied Question Answering with Global and Local Region Relevancy

Haochen Zhang, Nirav Savaliya, Faizan Siddiqui, Enna Sachdeva

Winter Conference on Applications of Computer Vision (WACV), 2026

FAST-EQA: Efficient Embodied Question Answering with Global and Local Region Relevancy.

PredMapNet: Future and Historical Reasoning for Consistent Online HD Vectorized Map Construction

Bo Lang, Nirav Savaliya, Zhihao Zheng, Jinglun Feng, Zheng-Hang Yeh, Mooi Choo Chuah

Winter Conference on Applications of Computer Vision (WACV), 2026

PredMapNet: Future and Historical Reasoning for Consistent Online HD Vectorized Map Construction.

TextFM: Robust Semi-dense Feature Matching with Language Guidance

Zhihao Zheng, Jinglun Feng, Nirav Savaliya, Zheng-Hang Yeh, Bo Lang, Mooi Choo Chuah

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026

TextFM: Robust Semi-dense Feature Matching with Language Guidance.

News

2026-02

Our work has been accepted at CVPR 2026!

2025-11

2 Papers have been accepted at WACV 2026!