Closed qjtang12 closed 5 years ago
What video are you using? Is it demo video or a video from your study? The direct cause to me is that the algorithm could not calculate the movement score from any pair of your video, possibly indicating that there is no salient feature/information to track with. It would be good if I can take a look at some demo frames of your video.
Download full resolution images Available until Apr 6, 2019 https://www.icloud.com/attachment/?u=https%3A%2F%2Fcvws.icloud-content.com%2FB%2FAYD3P_aaY4k30CeyPiMWuot1djc_AVyVuxB-yfPsDHiqumXtxDmUzBAH%2F%24%7Bf%7D%3Fo%3DAkODHKvFAaNNKc0G3saezFSivYiBSftLdCaI2A3ZsbHL%26v%3D1%26x%3D3%26a%3DB7S_RaZSJT38irFTc5ArQdYWbexUA5ZUSgEACAHIAP9SL6uVA9arSQ%26e%3D1554583062%26k%3D%24%7Buk%7D%26fl%3D%26r%3DF77896B5-CE3C-4D2A-8042-5D7BB328B0AD-1%26ckc%3Dcom.apple.largeattachment%26ckz%3D07CA434D-69DD-4458-9A41-22C0384EB4BF%26p%3D53%26s%3DtIRPNiax12pL5Ga-G8cEs-9mWSI&uk=BpsXbzjxFwxw95TPZZlQ6w&f=Images.zip&sz=42354962Hello Jinghao,
Thank you for getting back to me so quickly. I am use my own video recorded by UCLA miniscope. Regarding you explain, that sounds because my video quality is too bad. The reason why you see this shitty video is because my recording region is a little bit too deep. I am working on improving it too, if the script error can be solved by enhancing video quality. I have attached the video I ran yesterday.
Qijun Tang
On Mar 7, 2019, at 10:33 AM, JinghaoLu <notifications@github.com mailto:notifications@github.com> wrote:
What video are you using? Is it demo video or a video from your study? The direct cause to me is that the algorithm could not calculate the movement score from any pair of your video, possibly indicating that there is no salient feature/information to track with. It would be good if I can take a look at some demo frames of your video.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/JinghaoLu/MIN1PIPE/issues/15#issuecomment-470572701, or mute the thread https://github.com/notifications/unsubscribe-auth/AorOlDocVP_3zmYvuQnUCI4oPLAnV3m-ks5vUTFEgaJpZM4biETm.
https://github.githubassets.com/images/email/message_cards/header.png https://github.githubassets.com/images/email/message_cards/avatar.png https://github.com/JinghaoLu/MIN1PIPE https://github.com/JinghaoLu/MIN1PIPE/issues/15#issuecomment-470572701 http://schema.org/ https://github.com/JinghaoLu/MIN1PIPE/issues/15#issuecomment-470572701 https://github.com/JinghaoLu/MIN1PIPE/issues/15#issuecomment-470572701 https://github.com/
Hi Qijun,
I tested the video you provided, and it worked all well through the algorithm on my machine. I think improving imaging quality exactly solves the problem.
Hi Jinghao,
I am so sorry. I should have done this at very beginning but I just figured it out. The matlab on lab computer is lacking two packages. The problem could be either from the lack of Computer Vision System Toolbox or Curve Fitting Toolbox.
Thank you for helping me and thank you for your work.
Qijun On Mar 7, 2019, 4:27 PM -0500, JinghaoLu notifications@github.com, wrote:
Hi Qijun, I tested the video you provided, and it worked all well through the algorithm on my machine. I think improving imaging quality exactly solves the problem. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.
Hello Jinghao,
I am new to your package. When I try to test it out using demo. I got the Error as below.
Error using griddedInterpolant Interpolation requires at least two sample points in each dimension.
Error in interp1 (line 151) F = griddedInterpolant(X,V,method);
Error in get_trans_score (line 91) acorr(idss) = interp1(ids, acorr(ids), idss, 'pchip', 'extrap');
Error in frame_reg (line 80) acorr(max(1, idbatch(i) - 1): idbatch(i + 1) - 2) = get_trans_score(tmp, [], 1, 1, mq, maskc);
Error in min1pipe (line 102) [m, corr_score, raw_score, scl] = frame_reg(m, imaxy, se, Fsi_new, pixs, scl, sigma_x, sigma_f, sigma_d);