Presenting at ECCV 2022

European Conference on Computer Vision

October, 2022. Tel Aviv, Israel.

---

I was fortunate to have the opportunity to attend and present at the ECVA European Conference on Computer Vision (ECCV) in 2022.

ClearPose at the Main Conference Poster Session

We presented our work on ClearPose: Large-scale Transparent Object Dataset and Benchmark at the main conference poster session. Pictured below is myself with collaborator, Huijie Zhang. We appreciated the interest, questions, and discussion on the challenges of transparent object detection and pose estimation.

Talk at the Workshop on Recovering 6D Object Pose

In addition to ClearPose, our work on TransNet: Category-Level Transparent Object Pose Estimation was accepted to the 7th International Workshop on Recovering 6D Object Pose.

Below is a recording of our talk:

More ECCV Highlights

Attending ECCV is a unique opportunity to learn from state of the art computer vision research. From the University of Michigan, Samir Agarwala presented PlaneFormers: From Sparse View Planes to 3D Reconstruction. There were many interesting talks and posters at ECCV, including the work presented by Lin Huang and Dr. Tomáš Hodaň on Neural Correspondence Field for Object Pose Estimation and by Mathis Petrovich on TEMOS: Generating diverse human motions from textual descriptions.