1 / 6

Progress report

This report evaluates a parameter optimization strategy for a virtual camera system used in finger detection. The research identifies that optimal parameters represent 18% to 20% of all parameter combinations within a specific range. It emphasizes that finding these optimal parameters is quick, taking only a few minutes, yet notes limitations such as slow automatic search capabilities of existing TouchLib methods and lack of effectiveness under non-uniform lighting conditions. Potential improvements are proposed, including the integration of GPU acceleration for better performance and improved denoising methods.

mostyn
Télécharger la présentation

Progress report

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Progress report By shooting

  2. Problem with parameter optimization • Multi-optimal parameter • Optimal parameters are 18%~20% of all parameter combination through specific range • Only take few minutes for people to find an optimal parameter • Automatic search not fast enough

  3. Modified TouchLib • Disadvantage of TouchLib • Denoise method is not stable • Not suitable for non-uniform lighting condition • Without tracking and smooth mechanism

  4. Finger detection framework • Virtual camera • Intuitive in vision • Reduce matching problem • Unified finger size among different position • Detection module can be exchange • Upgrade performance (by implementing on GPU) IR cam Virtual Camera Detection module IR cam

  5. Fid 1 IR cam Detection module matching IR cam Detection module Fid 2 Fid 1 Same finger? Fid 2

  6. Parameter optimization • Enumeration • Parameter combination 6 (step) *6 (step) *6 (step) *6 (step) = 1296 • 244/1296 = 18.83% • Particle filter • Parameter combination • 50 (sample)*10 (iteration) = 500 • 108/500 = 21.6%

More Related