{"ID":14365,"CreatedAt":"2026-02-27T13:00:40Z","UpdatedAt":"2026-02-27T13:00:40Z","DeletedAt":null,"paper_url":"https://paperswithcode.com/paper/total-variation-based-dense-depth-from-multi","arxiv_id":"1711.07719","title":"Total Variation-Based Dense Depth from Multi-Camera Array","abstract":"Multi-Camera arrays are increasingly employed in both consumer and industrial\napplications, and various passive techniques are documented to estimate depth\nfrom such camera arrays. Current depth estimation methods provide useful\nestimations of depth in an imaged scene but are often impractical due to\nsignificant computational requirements. This paper presents a novel framework\nthat generates a high-quality continuous depth map from multi-camera\narray/light field cameras. The proposed framework utilizes analysis of the\nlocal Epipolar Plane Image (EPI) to initiate the depth estimation process. The\nestimated depth map is then processed using Total Variation (TV) minimization\nbased on the Fenchel-Rockafellar duality. Evaluation of this method based on a\nwell-known benchmark indicates that the proposed framework performs well in\nterms of accuracy when compared to the top-ranked depth estimation methods and\na baseline algorithm. The test dataset includes both photorealistic and\nnon-photorealistic scenes. Notably, the computational requirements required to\nachieve an equivalent accuracy are significantly reduced when compared to the\ntop algorithms. As a consequence, the proposed framework is suitable for\ndeployment in consumer and industrial applications.","url_abs":"http://arxiv.org/abs/1711.07719v1","url_pdf":"http://arxiv.org/pdf/1711.07719v1.pdf","authors":"[\"Hossein Javidnia\", \"Peter Corcoran\"]","published":"2017-11-21T00:00:00Z","tasks":"[\"Depth Estimation\"]","methods":"[]","has_code":false}