Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

cross-modal pseudo-labeling framework

A cross-modal pseudo-labeling framework is a machine learning system that automatically generates synthetic training annotations for one data modality by leveraging semantic information transferred from a complementary modality. In multimodal tasks such as open-vocabulary visual perception, the framework aligns shared representations across distinct data types, such as associating textual descriptions with corresponding visual features or spatial regions. These derived alignments produce pseudo-labels, including bounding boxes, segmentation masks, or category tags, without requiring exhaustive human annotations. The resulting pseudo-labels are then used to supervise or self-train downstream models on novel concepts, frequently utilizing filtering, confidence estimation, or distillation techniques to mitigate the adverse effects of label noise inherent in the automated generation process.

1 item

Open-Vocabulary Instance Segmentation via Robust Cross-Modal Pseudo-Labeling

Open-Vocabulary Instance Segmentation via Robust Cross-Modal Pseudo-Labeling

Dat Huynh, Jason Kuen, Zhe Lin, Jiuxiang Gu, Ehsan Elhamifar

OrganizationsAdobeNortheastern University

Why you should read this

Proposes a cross-modal pseudo-labeling framework that aligns caption words with visual mask features and filters label noise to segment novel object classes without mask annotations.

Open-vocabulary instance segmentation aims at segmenting novel classes without mask annotations. It is an important step toward reducing laborious human supervision. Most existing works first pretrain a model on captioned images covering many novel classes and then finetune it on limited base classes with mask annotations. However, the high-level textual information learned from caption pre-training alone cannot effectively encode the details required for pixel-wise segmentation. To address this, we propose a cross-modal pseudo-labeling framework, which generates training pseudo masks by aligning word semantics in captions with visual features of object masks in images. Thus, our framework is capable of labeling novel classes in captions via their word semantics to self-train a student model. To account for noises in pseudo masks, we design a robust student model that selectively distills mask knowledge by estimating the mask noise levels, hence mitigating the adverse impact of noisy pseudo masks. By extensive experiments, we show the effectiveness of our framework, where we significantly improve mAP score by 4.5% on MS-COCO and 5.1% on the large-scale Open Images & Conceptual Captions datasets compared to the state-of-the-art.

Added

2026-09-26