Daoyi Gao

I am a PhD student at the 3D AI Lab at Technical University of Munich.

Previously, I graduated from Robotics, Cognition, Intelligence at TUM. Prior to this, I obtained my B.E. from Beihang University (BUAA) majored in mechatronics.

My research interests are in the field of 3D computer vision, retrieval-based scene understanding.

Email  /  Google Scholar /  Github /  LinkedIn

profile photo
DiffCAD: Weakly-Supervised Probabilistic CAD Model Retrieval and Alignment from an RGB Image
Daoyi Gao, Dávid Rozenberszki, Stefan Leutenegger, Angela Dai
arXiv, 2023
project page / bibtex /

We proposed a weakly-supervised approach for CAD model retrieval and alignment from an RGB image. Our approach utilzes diffusion models to tackle the ambiguities in the monocular perception, and achives robuts cross-domain performance while only trained on synthetic dataset

Polarimetric Pose Prediction
Daoyi Gao*, Yitong Li*, Patrick Ruhkamp*, Iuliia Skobleva*, Magdalena Wysock*, HyunJun Jung, Pengyuan Wang, Arturo Guridi, Benjamin Busam
ECCV, 2022
project page / paper / bibtex / code

We proposed a hybrid model that utilizes polarizaiotn information with physical priors in a data-driven learning strategy to improve the accuracy of pose predictions for photometric challenging objects.

* Equal contribution. Alphabetical order.

Attention meets Geometry: Geometry Guided Spatial-Temporal Attention for Consistent Self-Supervised Monocular Depth Estimation
Patrick Ruhkamp*, Daoyi Gao*, Hanzhi Chen*, Nassir Navab, Benjamin Busam
3DV, 2021
project page / paper / bibtex / code

We proposed a self-supervised monocular depth estimation pipeline that aims to improve consistency while preserving accuracy. We also proposed a new Temporal Consistency Metric (TCM) to quantify depth consistency across frames.

* Equal contribution. Order determined randomly.

Spatial-Temporal Attention through Self-Supervised Geometric Guidance
Patrick Ruhkamp*, Daoyi Gao*, Hanzhi Chen*, Nassir Navab, Benjamin Busam
ICCV Workshop: Self-supervised Learning for Next-Generation Industry-level Autonomous Driving, 2021
project page / paper / code

We proposed a spatial-temporal attention mechanism guided by geometric constraints to aggregate geometrically meaningful attention and improve temporal depth stability and accuracy compared to previous methods.

* Equal contribution. Order determined randomly.

Teaching Assistant, Techniques in Artificial Intelligence (IN2062) WS 2021/2022

Teaching Assistant, Introduction to Deep Learning (IN2346) SS2020, WS 2020/2021

Teaching Assistant, Block Seminar Hands-on Deep Learning SS2020

Credicts: Jon Barron