Research
My research interests lie at the intersection of 3D computer vision and robotics. I am interested in building structured visual representations of the world to enable generalizable robot manipulation in open-world environments.
|
* Equal contribution
/
selected papers
|
|
DG16M: A Large-Scale Dataset for Dual-Arm Grasping with Force-Optimized Grasps
Md Faizal Karim*, Mohammed Saad Hashmi*, Shreya Bollimuntha, Mahesh Reddy Tapeti, Gaurav Singh, Nagamanikandan Govindan, K Madhava Krishna
IROS, 2025
project page
/
paper
A large-scale dataset of 16M dual-arm grasps with force-closure constraints, enabling better grasp synthesis for dual-arm manipulation.
|
|
SceneComplete: Open-World 3D Scene Completion in Complex Real World Environments for Robot Manipulation
Aditya Agarwal, Gaurav Singh, Bipasha Sen, Tomás Lozano-Pérez, Leslie Pack Kaelbling
Under Review, 2025
project page
/
arXiv
A novel pipeline for constructing complete, segmented 3D models from single views by composing pretrained perception modules.
|
|
DA-VIL: Adaptive Dual-Arm Manipulation with Reinforcement Learning and Variable Impedance Control
Md Faizal Karim*, Shreya Bollimuntha*, Mohammed Saad Hashmi, Autrio Das, Gaurav Singh, Srinath Sridhar, Arun Kumar Singh, Nagamanikandan Govindan, K Madhava Krishna
ICRA, 2025
project page
/
arXiv
A novel pipeline combining policy learning and gradient-based optimization for adaptive dual-arm manipulation with dynamic impedance control.
|
|
Constrained 6-DoF Grasp Generation on Complex Shapes for Improved Dual-Arm Manipulation
Gaurav Singh*, Sanket Kalwar*, Md Faizal Karim, Bipasha Sen, Nagamanikandan Govindan, Srinath Sridhar, K. Madhava Krishna
IROS, 2024
project page
/
arXiv
A diffusion-based grasp generative model that can generate dense grasps on target regions of complex objects without explicit constraint training.
|
|
HyP-NeRF: Learning Improved NeRF Priors using a HyperNetwork
Bipasha Sen*, Gaurav Singh*, Aditya Agarwal*, Rohith Agaram, K. Madhava Krishna, Srinath Sridhar
NeurIPS, 2023
project page
/
arXiv
A hypernetwork-based approach for learning generalizable category-level NeRF priors with improved quality and multi-view consistency.
|
|
SCARP: 3D Shape Completion in ARbitrary Poses for Improved Grasping
Bipasha Sen*, Aditya Agarwal*, Gaurav Singh*, B. Brojeshwar, Srinath Sridhar, K. Madhava Krishna
ICRA, 2023
project page
/
arXiv
A model that performs shape completion in arbitrary poses by learning disentangled pose and shape features, improving grasp proposals by 71.2%.
|
|