[R] CVPR'26 SPAR-3D Workshop Call For Papers

The organizing committee for the Computer Vision and Pattern Recognition (CVPR) 2026 conference has announced a deadline extension for its specialized SPAR-3D workshop, moving the final submission date to March 21, 2026. This workshop is dedicated to the critical and growing subfield of security, privacy, adversarial robustness, and reliability within 3D computer vision. The call is notably inclusive, welcoming not only papers wholly focused on these themes but also broader 3D vision contributions that contain a substantive discussion or dedicated section on robustness, safety, or trustworthiness. This reflects the increasing necessity of building these considerations directly into the development lifecycle of 3D models used in autonomous vehicles, robotics, and augmented reality.

The SPAR-3D workshop aims to consolidate cutting-edge research on vulnerabilities unique to 3D data and systems, such as adversarial attacks on point clouds, mesh reconstruction, or neural radiance fields (NeRFs). As 3D vision models become more integral to safety-critical applications, their resilience to manipulation, data poisoning, and privacy breaches is paramount. The extended deadline provides researchers additional time to prepare submissions that address these urgent challenges. The workshop will serve as a key forum for presenting novel defenses, evaluation benchmarks, and foundational theories, ultimately guiding the development of more secure and reliable 3D perception technologies. Submissions and further details are available on the official workshop website.