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

Fine-grained 3D shape retrieval aims to retrieve 3D shapes similar to a queryshape in a repository with models belonging to the same class, which requiresshape descriptors to be capable of representing detailed geometric informationto discriminate shapes with globally similar structures. Moreover, 3D objectscan be placed with arbitrary position and orientation in real-worldapplications, which further requires shape descriptors to be robust to rigidtransformations. The shape descriptions used in existing 3D shape retrievalsystems fail to meet the above two criteria. In this paper, we introduce anovel deep architecture, RISA-Net, which learns rotation invariant 3D shapedescriptors that are capable of encoding fine-grained geometric information andstructural information, and thus achieve accurate results on the task offine-grained 3D object retrieval. RISA-Net extracts a set of compact anddetailed geometric features part-wisely and discriminatively estimates thecontribution of each semantic part to shape representation. Furthermore, ourmethod is able to learn the importance of geometric and structural informationof all the parts when generating the final compact latent feature of a 3D shapefor fine-grained retrieval. We also build and publish a new 3D shape datasetwith sub-class labels for validating the performance of fine-grained 3D shaperetrieval methods. Qualitative and quantitative experiments show that ourRISA-Net outperforms state-of-the-art methods on the fine-grained objectretrieval task, demonstrating its capability in geometric detail extraction.The code and dataset are available at: https://github.com/IGLICT/RisaNET.

RISA-Net: Rotation-Invariant Structure-Aware Network for Fine-Grained 3D Shape Retrieval