Computer Science > Computer Vision and Pattern Recognition
[Submitted on 21 Apr 2024 (v1), last revised 11 Sep 2024 (this version, v6)]
Title:Attack on Scene Flow using Point Clouds
View PDF HTML (experimental)Abstract:Deep neural networks have made significant advancements in accurately estimating scene flow using point clouds, which is vital for many applications like video analysis, action recognition, and navigation. The robustness of these techniques, however, remains a concern, particularly in the face of adversarial attacks that have been proven to deceive state-of-the-art deep neural networks in many domains. Surprisingly, the robustness of scene flow networks against such attacks has not been thoroughly investigated. To address this problem, the proposed approach aims to bridge this gap by introducing adversarial white-box attacks specifically tailored for scene flow networks. Experimental results show that the generated adversarial examples obtain up to 33.7 relative degradation in average end-point error on the KITTI and FlyingThings3D datasets. The study also reveals the significant impact that attacks targeting point clouds in only one dimension or color channel have on average end-point error. Analyzing the success and failure of these attacks on the scene flow networks and their 2D optical flow network variants shows a higher vulnerability for the optical flow networks. Code is available at this https URL.
Submission history
From: Haniyeh Ehsani Oskouie [view email][v1] Sun, 21 Apr 2024 11:21:27 UTC (9,777 KB)
[v2] Sun, 28 Apr 2024 08:05:55 UTC (10,754 KB)
[v3] Tue, 18 Jun 2024 01:40:23 UTC (10,754 KB)
[v4] Sun, 25 Aug 2024 06:13:24 UTC (10,754 KB)
[v5] Tue, 27 Aug 2024 01:23:50 UTC (10,754 KB)
[v6] Wed, 11 Sep 2024 06:13:30 UTC (10,754 KB)
Current browse context:
cs.CV
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.