English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Advancing object detection to open-vocabulary and few-shot transfer has longbeen a challenge for computer vision research. This work explores a continuallearning approach that enables a detector to expand its zero/few-shotcapabilities via multi-dataset vision-language pre-training. Using naturallanguage as knowledge representation, we explore methods to accumulate "visualvocabulary" from different training datasets and unify the task as alanguage-conditioned detection framework. Specifically, we propose a novellanguage-aware detector OmDet and a novel training mechanism. The proposedmultimodal detection network can resolve the technical challenges inmulti-dataset joint training and it can generalize to arbitrary number oftraining datasets without the requirements for manual label taxonomy merging.Experiment results on COCO, Pascal VOC, and Wider Face/Pedestrian confirmed theefficacy by achieving on par or higher scores in joint training compared totraining separately. Moreover, we pre-train on more than 20 million images with4 million unique object vocabulary, and the resulting model is evaluated on 35downstream tasks of ODinW. Results show that OmDet is able to achieve thestate-of-the-art fine-tuned performance on ODinW. And analysis shows that byscaling up the proposed pre-training method, OmDet continues to improve itszero/few-shot tuning performance, suggesting a promising way for furtherscaling.

OmDet: Language-Aware Object Detection with Large-scale Vision-Language Multi-dataset Pre-training