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

Copy-Paste is a simple and effective data augmentation strategy for instancesegmentation. By randomly pasting object instances onto new background images,it creates new training data for free and significantly boosts the segmentationperformance, especially for rare object categories. Although diverse,high-quality object instances used in Copy-Paste result in more performancegain, previous works utilize object instances either from human-annotatedinstance segmentation datasets or rendered from 3D object models, and bothapproaches are too expensive to scale up to obtain good diversity. In thispaper, we revisit Copy-Paste at scale with the power of newly emerged zero-shotrecognition models (e.g., CLIP) and text2image models (e.g., StableDiffusion).We demonstrate for the first time that using a text2image model to generateimages or zero-shot recognition model to filter noisily crawled images fordifferent object categories is a feasible way to make Copy-Paste trulyscalable. To make such success happen, we design a data acquisition andprocessing framework, dubbed ``X-Paste", upon which a systematic study isconducted. On the LVIS dataset, X-Paste provides impressive improvements overthe strong baseline CenterNet2 with Swin-L as the backbone. Specifically, itarchives +2.6 box AP and +2.1 mask AP gains on all classes and even moresignificant gains with +6.8 box AP, +6.5 mask AP on long-tail classes. Our codeand models are available at https://github.com/yoctta/XPaste.

X-Paste: Revisiting Scalable Copy-Paste for Instance Segmentation using CLIP and StableDiffusion