CVPR 2021 Workshop

Frontiers of Monocular 3D Perception

Toyota Technological Institute at Chicago (TTIC) & Toyota Research Institute (TRI)

19th June 2021, Virtual

Overview

This workshop considers challenges and opportunities at the frontier of 3D reasoning from monocular visual input, the main modality common to most mobile phones, robots, and cars. Our main focus will be on 3D perception of scenes and objects directly from images and videos. The workshop’s main aim is to identify, characterize, and investigate how to qualitatively go beyond the current limitations in monocular 3D perception. The workshop consists of invited talks, panel discussions as well as a description of the Dense Depth for Autonomous Driving (DDAD) challenge and presentations from the winning methods. The total prize of the challenge is $5000.

Dense Depth for Autonomous Driving (DDAD) challenge

DDAD is a new autonomous driving benchmark from TRI (Toyota Research Institute) for long range and dense depth estimation in challenging and diverse urban conditions. DDAD includes high-resolution, long-range LiDAR sensors used to generate pointclouds and images from six calibrated cameras time-synchronized at 10 Hz. In addition to the raw data, DDAD also contains high-quality panoptic segmentation associated with a select set of validation and test frames.

The DDAD depth challenge consists of two tracks: self-supervised and semi-supervised monocular depth estimation. We will evaluate all methods against the ground truth Lidar depth, and we will also compute and report depth metric per semantic class. All evaluations will be done only on images from the front camera, and the winner will be chosen based on the abs_rel metric. In addition, a jury prize will be awarded to a 3rd competitor, based on the novelty of the method.

Program Details (tentative)

The timezone of the conference is GMT-5.

Speakers

Panelists

Organizers

The workshop is mainly organized by researchers from Toyota Technological Institute at Chicago (TTIC) & Toyota Research Institute (TRI).

Adrien Gaidon

ML Lead

TRI

Greg Shakhnarovich

Associate Professor

TTIC

Rares Ambrus

Research Scientist

TRI

Vitor Guizilini

Research Scientist

TRI

Igor Vasiljevic

Ph.D Student

TTIC

Matthew Walter

Assistant Professor

TTIC

Sudeep Pillai

Research Scientist

TRI

Nick Kolkin

Ph.D Student

TTIC

Contact

If you have any questions about the workshop, please contact us at rares.ambrus@tri.global / vitor.guizilini@tri.global / ivas@ttic.edu.