Result not same as image provided

[BukuBukuChagma]

So I don't know if I'm doing something wrong here or its suppose to be like this, when I provide a image in which I'm standing straight, in the body model it somehow translate to somewhat different from another angle. How to correct his? For example:

Why am I tilted? Why are the hands in not parallel to the body as in the image? Am I doing something wrong?

[fenghe12]

hello may i ask how did you use an image of a person as a input of smpl model and generate 3d representation？thanks！

[whl-007]

hello may i ask how did you use an image of a person as a input of smpl model and generate 3d representation？thanks！你好，请问你是如何使用一个人的图像作为SMPL模型的输入并生成3D表示的？谢谢！2谢谢！ Hello！Can you use own RGB images? I haven't found a way to use SMPLX to turn my own RGB images into SMPLX fits. Can you tell me where I should find the code?

[whl-007]

So I don't know if I'm doing something wrong here or its suppose to be like this, when I provide a image in which I'm standing straight, in the body model it somehow translate to somewhat different from another angle. How to correct his? For example:

Why am I tilted? Why are the hands in not parallel to the body as in the image? Am I doing something wrong?

So I don't know if I'm doing something wrong here or its suppose to be like this, when I provide a image in which I'm standing straight, in the body model it somehow translate to somewhat different from another angle. How to correct his? For example:

Why am I tilted? Why are the hands in not parallel to the body as in the image? Am I doing something wrong?

Hello！Can you use own RGB images? I haven't found a way to use SMPLX to turn my own RGB images into SMPLX fits. Can you tell me where I should find the code?

[BukuBukuChagma]

@whl-007 Clone the github repo of smplify-x. Install all the dependencies and requirements as mentioned in the repo. Once everything is setup, you can use the command mentioned in their README which im linking below in screenshot.

[whl-007]

@whl-007 Clone the github repo of smplify-x. Install all the dependencies and requirements as mentioned in the repo. Once everything is setup, you can use the command mentioned in their README which im linking below in screenshot.

"Hello, thank you for your response. So far, I have followed your instructions. I obtained the EHF.zip file, extracted it, and got the data for the image shown below. I used 01_img.jpg as the input and 01_2Djnt.json as the JSON file, but I encountered the following error: Traceback (most recent call last): File "/homesmplify-x-master/smplifyx/main-origin.py", line 290, in main(**args) File "/home/smplify-x-master/smplifyx/main-origin.py", line 246, in main fit_single_frame(img, keypoints[[person_id]], File "/home/smplify-x-master/smplifyx/fit_single_frame.py", line 120, in fit_single_frame assert (len(data_weights) == AssertionError: Number of Body pose prior weights 4 does not match the number of data term weights 5 Additionally, I am unsure how to generate the JSON file from my image. I used the OpenPose source code, but the JSON file I obtained does not match the format of the 01_2Djnt.json file provided. I hope you can help clarify my doubts. Thank you very much."

My json file show： {"keypoints": [[1449.0, 1469.0], [1534.0, 1737.0], [1272.0, 1737.0], [834.0, 1913.0], [482.0, 2002.0], [1711.0, 1737.0], [2150.0, 1825.0], [2588.0, 1913.0], [1535.0, 2625.0], [1359.0, 2626.0], [1273.0, 3161.0], [1272.0, 3695.0], [1710.0, 2626.0], [1624.0, 3161.0], [1624.0, 3695.0], [1448.0, 1468.0], [1536.0, 1468.0], [1360.0, 1469.0], [1623.0, 1469.0], [1624.0, 3961.0], [1711.0, 3874.0], [1623.0, 3783.0], [1184.0, 3874.0], [1097.0, 3873.0], [1359.0, 3695.0]]}

01_2Djnt.json file show：{"version":1.2,"people":[{"pose_keypoints_2d":[770.464,287.608,0.823277,790.065,334.916,0.671029,734.564,253.339,0.550223,695.431,214.197,0.563718,608.983,163.598,0.813454,847.126,385.5,0.662132,894.449,527.425,0.766871,781.892,507.847,0.729679,765.583,618.759,0.610898,708.468,610.604,0.556925,682.372,838.977,0.699029,679.115,1059.22,0.756149,825.938,626.921,0.562926,887.919,842.237,0.734691,943.375,1060.84,0.596494,759.043,251.71,0.844282,803.124,266.377,0.872308,739.472,240.287,0.247655,852.034,276.177,0.893368,959.706,1121.2,0.438599,977.649,1113.05,0.470517,938.491,1073.89,0.419639,631.811,1126.09,0.400877,622.021,1113.04,0.402617,693.792,1072.25,0.643139],"face_keypoints_2d":[738.952,247.045,0.089555,736.653,254.601,0.125376,734.353,264.456,0.142024,732.054,276.283,0.200632,734.682,291.395,0.230267,737.638,304.535,0.201998,745.194,332.459,0.278642,753.407,340.344,0.383774,763.92,344.943,0.332458,777.06,344.943,0.324587,792.172,337.715,0.432238,803.342,325.232,0.403625,815.168,315.705,0.423042,826.338,307.163,0.449559,832.908,297.308,0.488293,837.507,285.153,0.417282,840.464,270.698,0.294766,745.194,236.204,0.60619,752.422,236.532,0.742564,758.663,239.161,0.761214,765.891,243.76,0.724853,771.476,249.016,0.61468,785.602,251.973,0.581771,797.1,251.973,0.619972,807.941,252.958,0.664435,818.782,255.915,0.606848,827.652,262.814,0.61164,777.389,257.886,0.438152,775.746,268.07,0.339376,766.548,295.994,0.392756,764.577,302.893,0.486197,757.678,299.936,0.478495,760.306,305.192,0.58014,763.92,308.477,0.653194,769.176,308.806,0.511776,773.775,309.792,0.302411,750.451,249.673,0.71045,756.035,248.031,0.745266,762.277,251.644,0.671651,765.562,257.557,0.689411,758.663,257.886,0.758231,753.079,254.929,0.763183,795.129,267.413,0.701816,800.714,262.485,0.820026,808.269,263.142,0.714692,813.197,268.398,0.755492,807.941,270.698,0.730473,800.385,270.041,0.794726,748.151,310.449,0.400039,754.721,312.748,0.626283,760.635,314.391,0.6321,763.591,316.033,0.622115,768.19,315.705,0.564361,775.089,318.661,0.523215,783.631,320.304,0.251461,774.104,325.56,0.614065,767.862,326.217,0.706827,763.263,325.889,0.75772,758.335,325.56,0.768014,752.75,321.29,0.568902,750.122,313.405,0.521505,759.978,318.661,0.64907,763.591,319.319,0.634286,768.19,319.647,0.649023,780.346,320.633,0.3115,767.862,320.633,0.627485,763.591,320.961,0.738216,759.649,320.304,0.741162,758.335,252.301,0.682655,804.327,265.442,0.630357],"hand_left_keypoints_2d":[779.602,511.908,0.522801,752.64,492.837,0.538731,717.786,477.712,0.593758,690.824,472.451,0.645635,669.781,467.848,0.778695,705.949,519.141,0.804446,680.303,540.843,0.78207,663.205,556.625,0.799778,648.737,572.408,0.829955,717.129,539.527,0.747385,689.509,561.228,0.814414,668.466,578.326,0.777825,652.025,593.451,0.872193,731.596,552.022,0.773824,707.265,577.011,0.739739,688.852,592.136,0.727564,672.411,604.631,0.811094,742.775,565.174,0.724793,721.732,584.902,0.723509,709.237,596.082,0.779121,696.085,603.973,0.815069],"hand_right_keypoints_2d":[608.335,162.063,0.634607,594.151,163.531,0.777358,579.477,160.596,0.76749,565.781,154.237,0.811211,552.575,149.346,0.845225,582.411,124.889,0.750621,568.716,108.259,0.734522,555.02,96.0307,0.666811,544.259,89.672,0.600209,582.411,120.487,0.781987,565.292,101.9,0.762481,550.618,90.1611,0.725097,538.879,81.8459,0.748945,582.901,119.509,0.652711,566.27,101.9,0.585516,553.553,91.1394,0.55302,540.835,82.335,0.63448,582.411,119.02,0.478096,569.694,105.813,0.412964,562.846,97.9872,0.298483,553.553,90.6502,0.270408],"pose_keypoints_3d":[],"face_keypoints_3d":[],"hand_left_keypoints_3d":[],"hand_right_keypoints_3d":[]}]}

[whl-007]

I find some errors in smplifyx /fit_single_frame.py from line 108 to 120， data_weights‘s shape is 5，but body_pose_prior_weights‘s shape is 4 so that this will raise an error: AssertionError: Number of Body pose prior weights 4 does not match the number of data term weights 5 if data_weights is None: data_weights = [1, ] * 5

if body_pose_prior_weights is None:
    body_pose_prior_weights = [4.04 * 1e2, 4.04 * 1e2, 57.4, 4.78]

msg = (
    'Number of Body pose prior weights {}'.format(
        len(body_pose_prior_weights)) +
    ' does not match the number of data term weights {}'.format(
        len(data_weights)))
assert (len(data_weights) ==
        len(body_pose_prior_weights)), msg