RWTH Computer Vision Group

The poster for our #CoRL2023 paper on Lidar 4D panoptic segmentation will be up on Thursday (Nov 9) at 4:15pm! We propose a cool new multimodal approach for Lidar segmentation + tracking that leverages both Lidar scan and camera image data. Website: waabi.ai/4d-former/

RWTH Vision group at @CVPR!

💥 I'm pleased to announce our @CVPR Highlight paper: "TarViS: A Unified Approach for Target-based Video Segmentation". 💥TL;DR: We developed a single, jointly trained model that can tackle multiple video segmentation tasks e.g. VOS, VIS, VPS and achieve SOTA performance.

📢 Announcing Tracking & Its Many Guises (2nd Edition) workshop on June 18 at #CVPR2023! We are organizing two exciting multi-object tracking and segmentation challenges: Open-World & Long-Tail Tracking🏅Let’s innovate together!🤖 More info at: taodataset.org/workshop

Mask3D 🎭 is now at #ICRA2023, great work @JonasSchultCV! We use Mask Transformers for 3D Instance Segmentation on Point Clouds ~ 🥇 on ScanNet 📰Paper: arxiv.org/abs/2210.03105 🛠️Project: jonasschult.github.io/Mask3D/ 👨‍💻Code: github.com/JonasSchult/Ma… @Pandoro89 @orlitany @SiyuTang3

En route to #WACV2023! I'll present a paper on extreme multi-dataset learning of 3D human pose estimation when labels have different skeleton formats. Paper: arxiv.org/abs/2212.14474 Project page: istvansarandi.com/dozens

I'm pleased to announce our new dataset for object segmentation in video called BURST (Benchmark for Unified Recognition, Segmentation and Tracking). URL: github.com/Ali2500/BURST-… This'll be a game-changer for people working on VIS, VOS, MOTS & open-world tracking: (1/5)

I'll demo our latest 3D human pose model @eccvconf (Oct 26), trained on 28 different datasets at once. bit.ly/metrabs_demo youtu.be/sAF0atPmEec A lack of 3D pose data used to be a big issue, but many new datasets have been released in the last few years. Let's use them!

Our workshop paper on 'Differentiable Soft-Masked Attention' is now up on Arxiv: arxiv.org/pdf/2206.00182… We'll be presenting this at the Transformers for Vision (T4V) workshop at @CVPR'22!

Pleased to announce that we have 2x #CVPR2022 oral papers accepted: 1. HODOR (yes that's the name of the paper): arxiv.org/abs/2112.09131 2. Opening up Open World Tracking arxiv.org/pdf/2104.11221… @aliathar94 @JonathonLuiten @Pandoro89 @AljosaOsep

MeTRAbs got a new EffNetV2 backbone and works better than ever. Simple API, everything packed into one easy-to-use TensorFlow SavedModel. No repo cloning, just get the model and predict. Code, model links: github.com/isarandi/metra… More qualitative results: youtu.be/4VFKiiW9RCQ

Your 3D model is overfitting?🥴 Then check out Mix3D!🤗 Simply "mixing" two 3D scenes improves generalization in only 4 lines of code!🤓🤯 Great work! @kumuji @JonasSchultCV @orlitany 🥇 1st Place on ScanNet 🤩 Oral @3DVconf 📰 arxiv.org/abs/2110.02210 🛠️ nekrasov.dev/mix3d

Today @CVPR visit our paper "From Points to Multi-Object 3D Reconstruction" Joint reconstruction of multiple objects🍣 from a monocular image🧐 @krematas B.Leibe @VittoFerrariCV 👉#CVPR2021 Session 4 ⏰17:00 CET / 11:00 EDT 📄arxiv.org/abs/2012.11575 🧑‍💻francisengelmann.github.io/points2objects

Multi-Object Tracking (MOT) has been notoriously difficult to evaluate, and evaluation has been a constant source of frustration for many. Check out this blog post (jonathonluiten.medium.com/how-to-evaluat… ) which describes our recent work on the HOTA metrics for better tracking evaluation! 1/6

The TensorFlow implementation of our MeTRAbs pose estimator just got highlighted on #TFCommunitySpotlight!

DR-SPAAM, our real-time person detector from 2D range data, has been picked by @NVIDIAEmbedded as the "Jetson Project of the Month" for September 2020! #dr_spaam" target="_blank" rel="nofollow noopener">developer.nvidia.com/embedded/commu… Paper (IROS'20): arxiv.org/abs/2004.14079 Code: github.com/VisualComputin… Video: youtube.com/watch?v=fACppM…

Our state-of-the-art MeTRAbs absolute 3D human pose estimation model is now released for #TensorFlow! @IEEEFG2020 @RWTHVisionLab Code: github.com/isarandi/metra… Video: youtu.be/BemM8-Lx47g Paper: arxiv.org/abs/2007.07227

Check out our WACV'21 paper on "Reducing the Annotation Effort for Video Object Segmentation Datasets", we introduce a new video object segmentation benchmark! arxiv.org/abs/2011.01142 #wacv #wacv2021 #wacv21 #Videoobjectsegmentation

Tomorrow we'll be having live Q/A sessions for our paper "Making a Case for 3D Convolutions for Object Segmentation in Videos" at BMVC 2020: bmvc2020-conference.com/conference/pap… @RWTHVisionLab @AljosaOsep @lealtaixe

Join the Q/A session tomorrow at ECCV'20 for our paper "STEm-Seg: Spatio-temporal Embeddings for Instance Segmentation in Videos". We'll be on via Zoom at 06:00-08:00 and 14:00-16:00 (UTC+1). @eccvconf @RWTHVisionLab @ComputerVision