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

Instance segmentation is a fundamental research in computer vision,especially in autonomous driving. However, manual mask annotation for instancesegmentation is quite time-consuming and costly. To address this problem, someprior works attempt to apply weakly supervised manner by exploring 2D or 3Dboxes. However, no one has ever successfully segmented 2D and 3D instancessimultaneously by only using 2D box annotations, which could further reduce theannotation cost by an order of magnitude. Thus, we propose a novel frameworkcalled Multimodal Weakly Supervised Instance Segmentation (MWSIS), whichincorporates various fine-grained label generation and correction modules forboth 2D and 3D modalities to improve the quality of pseudo labels, along with anew multimodal cross-supervision approach, named Consistency Sparse Cross-modalSupervision (CSCS), to reduce the inconsistency of multimodal predictions byresponse distillation. Particularly, transferring the 3D backbone to downstreamtasks not only improves the performance of the 3D detectors, but alsooutperforms fully supervised instance segmentation with only 5% fullysupervised annotations. On the Waymo dataset, the proposed frameworkdemonstrates significant improvements over the baseline, especially achieving2.59% mAP and 12.75% mAP increases for 2D and 3D instance segmentation tasks,respectively. The code is available athttps://github.com/jiangxb98/mwsis-plugin.

MWSIS: Multimodal Weakly Supervised Instance Segmentation with 2D Box Annotations for Autonomous Driving