Instance-Level Recognition Workshop at ECCV'24

Workshop Location: Amber 5

Sep. 30th, 9:00am-12:40pm (UTC+2/CEST)

ILR2024

Our workshop is focused on visual Instance-Level Recognition (ILR), with a primary objective of identifying, comparing, or synthesizing images related to specific objects, scenes, or events. Unlike the broad categorization found in category-level recognition, where classes are defined semantically (e.g., "a chair"), ILR delves into tasks with the utmost granularity in class definition, such as identifying "the chair of my desk".

This year, we expand the scope of our workshop by introducing a call for papers, in addition to hosting keynote talks by renowned speakers and invited paper talks from the main conference.

The 2024 Instance-Level Recognition (ILR) Workshop is a follow-up of five successful editions of our previous workshops — the first two having focused only on landmark recognition (CVPRW18, CVPRW19), the following ones expanding to the domains of artworks and products (ECCVW20, ICCVW21), and the latest one introducing the universal image embedding problem (ECCVW22).

Workshop Schedule

Welcome Remarks

Slides

Sep. 30th, 9:00am-9:10am (UTC+2/CEST)

Keynote 1

Cordelia Schmid
Fine-grained image classification based on retrieval and data generation

Sep. 30th, 9:10am-9:40am (UTC+2/CEST)

Oral session

  • Segment, Select, Correct: A Framework for Weakly-Supervised Referring Segmentation
  • Leveraging Object Priors for Point Tracking
Sep. 30th, 9:40am-10:05am (UTC+2/CEST)

Poster session & Coffee break

Sep. 30th, 10:05am-11:05am (UTC+2/CEST)

Keynote 2

Giorgos Kordopatis-Zilos
Visual similarity learning for instance-level image and video retrieval

Slides

Sep. 30th, 11:05am-11:35am (UTC+2/CEST)

Oral Session - Invited papers

  • Grounding Language Models for Visual Entity Recognition
  • PetFace: A Large-Scale Dataset and Benchmark for Animal Identification
  • MeshVPR: Citywide Visual Place Recognition Using 3D Meshes
Sep. 30th, 11:35am-12:05pm (UTC+2/CEST)

Keynote 3

Varun Jampani
Instance-specific 2D and 3D generation

Slides

Sep. 30th, 12:05pm-12:35pm (UTC+2/CEST)

Closing Remarks

Sep. 30th, 12:35pm-12:40pm (UTC+2/CEST)

Keynote Speakers

Cordelia Schmid

Research Director at INRIA and Google DeepMind

Fine-grained image classification based on retrieval and data generation

Varun Jampani

Lead Researcher at Stability AI

Instance-specific 2D and 3D generation

Giorgos Kordopatis-Zilos

Postdoctoral Researcher at the CTU in Prague

Visual similarity learning for instance-level image and video retrieval


Accepted Papers

Long Papers

  • Segment, Select, Correct: A Framework for Weakly-Supervised Referring Segmentation (Oral)
    Francisco Eiras, Kemal Oksuz, Adel Bibi, Philip Torr, Puneet K. Dokania
  • Leveraging Object Priors for Point Tracking (Oral)
    Bikram Boote, Ngoc Anh Thai, Wenqi Jia, Ozgur Kara, Stefan Stojanov, James Matthew Rehg, Sangmin Lee

    • Short Papers

  • Image Re-ranking with Long-Context Sequence Modeling
    Zilin Xiao, Ayush Sachdeva, Hao-Jen Wang, Vicente Ordonez
  • Promptable Iterative Visual Refinement for Video Instance Segmentation
    Tuyen Tran, Thao Minh Le, Truyen Tran
  • Continual Hyperbolic Learning of Instances and Classes
    Melika Ayoughi, Mina Ghadimi Atigh, Mohammad Mahdi Derakhshani, Cees G. M. Snoek, Pascal Mettes, Paul Groth

    • Invited Papers

  • Grounding Language Models for Visual Entity Recognition (Oral)
    Zilin Xiao, Ming Gong, Paola Cascante-Bonilla, Xingyao Zhang, Jie Wu, Vicente Ordonez
  • PetFace: A Large-Scale Dataset and Benchmark for Animal Identification (Oral)
    Risa Shinoda, Kaede Shiohara
  • MeshVPR: Citywide Visual Place Recognition Using 3D Meshes (Oral)
    Gabriele Berton, Lorenz Junglas, Riccardo Zaccone, Thomas Pollok, Barbara Caputo, Carlo Masone
  • AMES: Asymmetric and Memory-Efficient Similarity Estimation for Instance-level Retrieval
    Pavel Suma, Giorgos Kordopatis-Zilos, Ahmet Iscen, Giorgos Tolias
  • MarineInst: A Foundation Model for Marine Image Analysis with Instance Visual Description
    Ziqiang Zheng, Yiwei Chen, Huimin Zeng, Tuan-Anh Vu, Binh-Son Hua, Sai Kit Yeung
  • "Where am I?" Scene Retrieval with Language
    Jiaqi Chen, Daniel Barath, Iro Armeni, Marc Pollefeys, Hermann Blum
  • Revisit Anything: Visual Place Recognition via Image Segment Retrieval
    Kartik Garg, Sai Shubodh Puligilla, Shishir N Y Kolathaya, Madhava Krishna, Sourav Garg
  • Powerful and Flexible: Personalized Text-to-Image Generation via Reinforcement Learning
    Fanyue Wei, Wei Zeng, Zhenyang Li, Dawei Yin, Lixin Duan, Wen Li
  • On Learning Discriminative Features from Synthesized Data for Self-Supervised Fine-Grained Visual Recognition
    Zihu Wang, Lingqiao Liu, Scott Ricardo Figueroa Weston, Samuel Tian, Peng Li
  • Open-Set Recognition in the Age of Vision-Language Models
    Dimity Miller, Niko Suenderhauf, Alex Kenna, Keita Mason

    • Call For Papers

    We call for novel and unpublished work in the format of long papers (14 pages excluding references) and short papers (4 pages excluding references). Papers should follow the ECCV proceedings style and will be reviewed in a double-blind fashion. Accepted papers will be presented at the workshop either as a poster or as an oral talk. Only long papers will be published in the ECCV workshop proceedings. All submissions will be handled electronically via the OpenReview conference submission website.

    Topics of interest include

  • particular object/event retrieval
  • instance-level object classification, detection and pose estimation
  • image matching and video tracking
  • instance-level image generation
  • other ILR applications
  • universal feature learning for instance-level recognition
  • challenges in ILR, such has long-tail distribution and open vocabulary
  • evaluation of large-scale ILR system
  • ILR datasets
  • leveraging Vision-Language-Models (VLM) to solve ILR problems
  • visual geo-localization

    • The task of person re-identification clearly falls within our definition of ILR. Nevertheless, because of its social implications, we intentionally omit it from the list of topics.

    Important Dates

  • Submission deadline: July 25th, 2024
  • Paper notification: August 8th, 2024
  • Camera-ready deadline: August 15th, 2024
  • Workshop date: September 30th, 2024

    • Questions? Please reach out to us at ilr-workshop@googlegroups.com

    Organizers

    Andre Araujo

    Google DeepMind (Primary Contact)

    Bingyi Cao

    Google DeepMind

    Kaifeng Chen

    Google DeepMind

    Ondrej Chum

    Czech Technical University

    Noa Garcia

    Osaka University

    Bohyung Han

    Seoul National University

    Guangxing Han

    Columbia University

    Giorgos Tolias

    Czech Technical University

    Hao Yang

    Amazon

    Nikolaos-Antonios Ypsilantis

    Czech Technical University

    Xu Zhang

    Amazon

    © 2024 ILR2024

    We thank Jalpc for the jekyll template