search

cfzd / Ultra-Fast-Lane-Detection-v2

Ultra Fast Deep Lane Detection With Hybrid Anchor Driven Ordinal Classification (TPAMI 2022)
MIT License
607 stars 100 forks source link

自定义Tusimple格式数据集 #143

Open hwang12345 opened 1 year ago

hwang12345 commented 1 year ago

自己标注1920*1080分辨率的图像，标注后整理为Tusimple数据集格式，并且对h_samples进行了修改，具体如下： [135, 152, 169, 186, 203, 220, 237, 254, 271, 288, 305, 322, 339, 356, 373, 390, 407, 424, 441, 458, 475, 492, 509, 526, 543, 560, 577, 594, 611, 628, 645, 662, 679, 696, 713, 730, 747, 764, 781, 798, 815, 832, 849, 866, 883, 900, 917, 934, 951, 968, 985, 1002, 1019, 1036, 1053, 1070] 并且修改了label = np.zeros((1080, 1920), dtype=np.uint8) 将720 ，1280 修改为1080 1920, yy = ((np.array(lines[which_lane][1::2]) - 160 ) / 10).astype(int) 将160 修改为135,程序运行报错：yy超出范围56。 我的修改是有问题的吗？

File "convert_tusimple.py", line 217, in generate_segmentation_and_train_list(args.root, line_txt, names) File "convert_tusimple.py", line 171, in generate_segmentation_and_train_list all_points[2, yy, 0] = xx IndexError: index 56 is out of bounds for axis 1 with size 56

GIN-zhao commented 1 year ago

项目中设置了范围

hustophi commented 9 months ago

According to your error info, you should modify the file 'convert_tusimple.py',and the follow is which part you shoud modify according to yourselfs data.

label = np.zeros((720,1280),dtype=np.uint8)
all_points = np.zeros((4,56,2), dtype=np.float)
the_anno_row_anchor = np.array([160, 170, 180, 190, 200, 210, 220, 230, 240, 250, 260,
       270, 280, 290, 300, 310, 320, 330, 340, 350, 360, 370,
       380, 390, 400, 410, 420, 430, 440, 450, 460, 470, 480,
       490, 500, 510, 520, 530, 540, 550, 560, 570, 580, 590,
       600, 610, 620, 630, 640, 650, 660, 670, 680, 690, 700,
       710])

GOOD LUCK!

GIN-zhao commented 9 months ago

Thank you, I have already resolved it before.hahaha

---原始邮件--- 发件人: @.> 发送时间: 2024年1月22日(周一) 凌晨0:52 收件人: @.>; 抄送: @.**@.>; 主题: Re: [cfzd/Ultra-Fast-Lane-Detection-v2] 自定义Tusimple格式数据集 (Issue #143)

According to your error info, you should modify the file 'convert_tusimple.py',and the follow is which part you shoud modify according to yourselfs data. label = np.zeros((720,1280),dtype=np.uint8) all_points = np.zeros((4,56,2), dtype=np.float) the_anno_row_anchor = np.array([160, 170, 180, 190, 200, 210, 220, 230, 240, 250, 260, 270, 280, 290, 300, 310, 320, 330, 340, 350, 360, 370, 380, 390, 400, 410, 420, 430, 440, 450, 460, 470, 480, 490, 500, 510, 520, 530, 540, 550, 560, 570, 580, 590, 600, 610, 620, 630, 640, 650, 660, 670, 680, 690, 700, 710])

GOOD LUCK!

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.***>