{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/direct-visual-odometry-using-bit-planes","title":"Direct Visual Odometry using Bit-Planes","arxiv_id":"1604.00990","date":"2016-04-04","proceeding":null,"authors":["Hatem Alismail","Brett Browning","Simon Lucey"],"abstract":"Feature descriptors, such as SIFT and ORB, are well-known for their\nrobustness to illumination changes, which has made them popular for\nfeature-based VSLAM\\@. However, in degraded imaging conditions such as low\nlight, low texture, blur and specular reflections, feature extraction is often\nunreliable. In contrast, direct VSLAM methods which estimate the camera pose by\nminimizing the photometric error using raw pixel intensities are often more\nrobust to low textured environments and blur. Nonetheless, at the core of\ndirect VSLAM is the reliance on a consistent photometric appearance across\nimages, otherwise known as the brightness constancy assumption. Unfortunately,\nbrightness constancy seldom holds in real world applications.\n  In this work, we overcome brightness constancy by incorporating feature\ndescriptors into a direct visual odometry framework. This combination results\nin an efficient algorithm that combines the strength of both feature-based\nalgorithms and direct methods. Namely, we achieve robustness to arbitrary\nphotometric variations while operating in low-textured and poorly lit\nenvironments. Our approach utilizes an efficient binary descriptor, which we\ncall Bit-Planes, and show how it can be used in the gradient-based optimization\nrequired by direct methods. Moreover, we show that the squared Euclidean\ndistance between Bit-Planes is equivalent to the Hamming distance. Hence, the\ndescriptor may be used in least squares optimization without sacrificing its\nphotometric invariance. Finally, we present empirical results that demonstrate\nthe robustness of the approach in poorly lit underground environments.","url_abs":"http://arxiv.org/abs/1604.00990v1","url_pdf":"http://arxiv.org/pdf/1604.00990v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"direct-visual-odometry-using-bit-planes","repo_url":"https://github.com/wccdyp/bpvo","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"visual-odometry","task_name":"Visual Odometry"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}