Christos Sakaridis

Monocular depth estimation is “impossible” because one image can’t measure depth geometrically. Our iDisc #CVPR2023 can group pixels w/o supervision and learn depth inductive bias on groups. We get LiDAR-like (but denser) depth from single images! More: vis.xyz/pub/idisc

I am pleased to give a talk at the #ICLR2023 Scene Representations for Autonomous Driving Workshop on "Optimizing internal network representations for geometric and semantic perception" and present our latest works in this direction iclr.cc/virtual/2023/w… live.bilibili.com/27720575

Check out our #CVPR2022 paper "P3Depth: Monocular Depth Estimation with a Piecewise Planarity Prior". Paper: openaccess.thecvf.com/content/CVPR20… Code: github.com/SysCV/P3Depth Poster @ CVPR: 21/6, 10:00AM - 12:30PM (CST), Halls B2-C My co-authors: Vaishakh Patil, @alexliniger, Luc Van Gool

I am happy to announce that the code for our @CVPR paper "LiDAR Snowfall Simulation for Robust 3D Object Detection" is now publicly available for non-commercial use: github.com/SysCV/LiDAR_sn… #LiDAR #snowfall #simulation #oral: Friday, June 24th at 10:10 AM (CST) in Great Hall B-C

CVPR 2022: Call for Workshop Paper Sign up: git.io/SR22

ACDC Challenge 2022 #challenge2022" target="_blank" rel="nofollow noopener">acdc.vision.ee.ethz.ch/news#challenge… on semantic segmentation in adverse conditions is up! The challenge is held together with the CVPR'22 workshop Vision for All Seasons, it is based on the public ACDC benchmarks acdc.vision.ee.ethz.ch/benchmarks and it will run until Jun 3, 2022.

Vision for All Seasons: Adverse Weather and Lighting Conditions 7th New Trends in Image Restoration and Enhancement Workshop and Challenges Computer Vision for Fashion, Art, and Design

Is there already a final decision on whether CVPR 2022 will be hybrid or virtual? @CVPR

Has the NeurIPS 2022 paper submission deadline been announced or not yet?

Poster sessions for "Fog Simulation on Real LiDAR Point Clouds for 3D Object Detection in Adverse Weather": Wed. Oct. 13th, 7:00 PM EDT / Thu. Oct. 14th 1:00 AM CEST / 7:00 AM CST + Fri. Oct. 15th, 12:00 PM EDT / 6:00 PM CEST / Sat. Oct. 16th, 12:00 AM CST

Poster sessions for "ACDC: The Adverse Conditions Dataset with Correspondences for Semantic Driving Scene Understanding": Wed. Oct. 13th, 9:00 AM EDT / 3:00 PM CEST / 9:00 PM CST + Fri. Oct. 15th, 5:00 PM EDT / 11:00 PM CEST / Sat. Oct. 16th, 5:00 AM CST

You are welcome to visit our posters at #ICCV2021 !

2 papers accepted to #ICCV2021! Paper 1: ACDC dataset, with Dengxin Dai and Luc Van Gool. See also: acdc.vision.ee.ethz.ch Paper 2: Fog simulation for LiDAR-based 3D object detection, with Martin Hahner, Dengxin Dai and Luc Van Gool.

I will present our latest #AISTATS work about sparse and stable CNN parameterizations (pinned) in the sparsity workshop today in 4 hrs #SNN21

I am happy to give a talk about our new ACDC dataset in the #IV2021 workshop on Autonomy@Scale on July 11. ki-deltalearning.de/event?tx_news_…

PCs update: Thanks for your patience. 99.8% of all papers have 3 reviews completed. We have 10 papers still to chase today. We will release of the reviews **around 17:00 PST**, so please don't keep refreshing CMT until then. We'll announce on Twitter as promised.

Glad to share one of the accepted papers at CVPR2021: - Learning to Relate Depth and Semantics for Unsupervised Domain Adaptation - arxiv.org/pdf/2105.07830… Suman Saha, Anton Obukhov, Danda Pani Paudel, Menelaos Kanakis, Yuhua Chen, Stamatios Georgoulis, Luc Van Gool

@AntonObukhov1 Thanks Anton!

My colleague @CSakaridis, who works on making self-driving cars safer to use in harsh weather conditions, has just released a #cityscapes-inspired benchmark, a paper, and joined Twitter. Welcome Christos!

ACDC released: a new large-scale driving dataset for training and testing semantic segmentation algorithms on adverse visual conditions, such as fog, nighttime, rain, and snow. The dataset and benchmarks are available at acdc.vision.ee.ethz.ch #computervision #autonomouscars