first-author

Denoising Vision Transformers

ECCV 2024 Investigating positional embedding artifacts in ViTs, and denosing them.

Augmenting Lane Perception and Topology Understanding with Standard Definition Navigation Maps

ICRA 2024 The first work to investigate how to leverage standard definition maps to augment lane topology reasoning.

Better Monocular 3D Detectors with LiDAR from the Past

ICRA 2024 Developed a method that uses repeated traversals to improve 3D object detection from camera inputs.

Pre-training LiDAR-based 3D Object Detectors through Colorization

ICLR 2024 Self-supervised method of pre-training LiDAR based object detection leveraging camera color information.

Reward Finetuning for Faster and More Accurate Unsupervised Object Discovery

NeurIPS 2023 Discovering 3D objects with reward fine-tuning, drawing inspiration from the RL community.

Unsupervised Adaptation from Repeated Traversals for Autonomous Driving

NeurIPS 2022 Domain adaptation method for 3D object detection leveraging repeated traversals.

Learning to Detect Mobile Objects from LiDAR Scans Without Labels

CVPR 2022 Discovering 3D objects from repeated traversals and self training.

Stay Positive: Non-Negative Image Synthesis for Augmented Reality

CVPR 2021  __(Oral)__ Solving non-negative image generation for Augmented Reality.

Safety-Oriented Pedestrian Motion and Scene Occupancy Forecasting

IROS 2021 Pedestrian motion forecasting with a focus on safety.