Re:Cognize -- Open-Set Comic Character Re-Identification
cs.CV, cs.MM
Submitted: 2026-09-27
Updated: 2026-09-27
Comments: Accepted at NeurIPS 2026 ED Track
Code: https://github.com/eternal-f1ame/Re-Cognize
Project page: https://re-cognize.vercel
License: http://creativecommons.org/licenses/by/4.0/
The gist: A manga reader meets a character on one page and knows them on sight a hundred pages later, without ever being handed a cast list.
Terminology
Abstract
A manga reader meets a character on one page and knows them on sight a hundred pages later, without ever being handed a cast list. Re-identifying comic characters demands the same, open-set and sequential: pages arrive as a stream in reading order, new faces appear before anyone names them, and the cast is assembled as the story is read. Re:Cognize evaluates recognition as the story is read, not against a cast handed over in advance: four protocols on one query stream, from closed-set retrieval to a cast the model must build and grow itself. The surprise is where models fail. Recognising is close to solved: one reference image per character already ranks as well as a gallery built in advance. Knowing what to believe is not: a model that adds its own matches makes its cast worse, while the same growth with correct labels would gain over twenty points of top-1 accuracy. The bottleneck is acceptance, not vision, and one comparison decides it: an addition pays exactly when it is right more often than the cast already was on the queries it takes over. The comparison has nothing to fit, and measured on half of a new corpus it calls the other half correctly. ReCast puts it to work with nothing fitted on data: a cast sheet of one running average per character, grown only where the page itself vouches for a crop. It recovers a third to two thirds of what perfect labels would, depending on whether the cast starts from random examples or from first appearances. Re:Cognize measures whether a model can read along; ReCast is a cast that does. Our claims are on identity maintenance, recognising characters already met; the emergence of new ones is measured as a diagnostic under a fixed reference rule, and we propose no method for it.
Sources
- Re:Verse -- Can Your VLM Read a Manga?
- In Defense of the Triplet Loss for Person Re-Identification
- Distribution-aware Forgetting Compensation for Exemplar-Free Lifelong Person Re-identification
- Object Detection for Comics using Manga109 Annotations
- Identity-Aware Semi-Supervised Learning for Comic Character Re-Identification
- One missing piece in Vision and Language: A Survey on Comics Understanding
- Deep Recurrent Convolutional Networks for Video-based Person Re-identification: An End-to-End Approach
- Unsupervised Manga Character Re-identification via Face-body and Spatial-temporal Associated Clustering
- ReID5o: Achieving Omni Multi-modal Person Re-identification in a Single Model
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models