[visionlist] [CfP] 8th Instance-Level Recognition and Generation Workshop @ ECCV2026

[Giorgos Kordopatis-Zilos]

[Apologies if you receive multiple copies]

Call for Papers – ILR+G at ECCV2026

==============================================================
8th Instance-Level Recognition and Generation (ILR+G) Workshop
European Conference on Computer Vision, ECCV 2026
Malmö, Sweden, September 8-12th, 2026
https://ilr-workshop.github.io/ECCVW2026
==============================================================

Relevant topics
---------------
The Instance-Level Recognition and Generation (ILR+G) workshop aims to 
explore computer vision tasks focusing on specific instances rather than 
broad categories, covering both recognition (instance-level recognition 
- ILR) and generation (instance-level generation - ILG). Instance-level 
(IL) represents the finest granularity of visual understanding, where 
each unique object, scene, or event constitutes its own class. This 
contrasts with semantic-level (SL), which relies on broader class 
definitions based on general semantic content. For example, under SL, an 
artwork image may be labeled as (e.g.) “a painting”, whereas under IL, 
it is identified as a specific work, (e.g.) “Blue Poles” by Jackson 
Pollock. Two objects share a class label only if they are visually 
indistinguishable, naturally leading to open-world settings with vast 
numbers of classes and increased challenges. While ILR focuses on 
distinguishing specific instances, ILG aims to synthesize images or 
videos of a particular object while preserving its identity. Despite its 
importance in real-world applications, IL has received less attention 
than SL. This workshop aims to highlight core challenges, diverse 
applications, and common ground across ILR and ILG, while fostering 
exploration of potential synergies.

Relevant topics include (but are not limited to):
- instance-level object classification, detection, segmentation, and 
pose estimation
- particular object and event retrieval
- personalized image and video generation
- cross-modal/multi-modal recognition at instance-level
- other ILR tasks such as image matching, visual geo-localization, 
animal re-identification, copy detection, video tracking, moment retrieval
- other ILR+G applications, datasets, and benchmarks

Even though tasks such as person and vehicle re-identification fall 
within the definition of ILR, we intentionally omit them from the list 
of topics, due to ethical and social implications. Submitted papers on 
those topics will be desk rejected.

submission details
------------------
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 
undergo double-blind peer review. Selected long papers will be invited 
for oral presentations; all accepted papers will be presented as 
posters. Only long papers will be published in the ECCV workshop 
proceedings. This year, financial awards will be given to the best 
papers, and student support grants will be provided to eligible 
participants. All submissions will be handled electronically via the 
OpenReview submission system.


important dates
---------------
submission deadline: June 26, 2026
notification of acceptance: July 24, 2026
camera-ready papers due: August 15, 2026


invited speakers
----------------
Georgia Gkioxari, Caltech
Adam Harley, Meta Reality Labs
Richard Zhang, Adobe Research (TBC)


organising committee
--------------------
Andre Araujo, Google DeepMind
Bingyi Cao, Google DeepMind
Kaifeng Chen, xAI
Ondrej Chum, Czech Technical University in Prague
Noa Garcia, University of Osaka
Guangxing Han, Google DeepMind
Giorgos Kordopatis-Zilos, Czech Technical University in Prague
Giorgos Tolias, Czech Technical University in Prague
Yankun Wu, University of Osaka
Hao Yang, Amazon
Nikolaos-Antonios Ypsilantis, Czech Technical University in Prague
Xu Zhang, Amazon