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

For saving cost, many deep neural networks (DNNs) are trained on third-partydatasets downloaded from internet, which enables attacker to implant backdoorinto DNNs. In 2D domain, inherent structures of different image formats aresimilar. Hence, backdoor attack designed for one image format will suite forothers. However, when it comes to 3D world, there is a huge disparity amongdifferent 3D data structures. As a result, backdoor pattern designed for onecertain 3D data structure will be disable for other data structures of the same3D scene. Therefore, this paper designs a uniform backdoor pattern: NRBdoor(Noisy Rotation Backdoor) which is able to adapt for heterogeneous 3D datastructures. Specifically, we start from the unit rotation and then search forthe optimal pattern by noise generation and selection process. The proposedNRBdoor is natural and imperceptible, since rotation is a common operationwhich usually contains noise due to both the miss match between a pair ofpoints and the sensor calibration error for real-world 3D scene. Extensiveexperiments on 3D mesh and point cloud show that the proposed NRBdoor achievesstate-of-the-art performance, with negligible shape variation.

Be Careful with Rotation: A Uniform Backdoor Pattern for 3D Shape