Scannet v2

[ghost]

Hello Your work is outstanding! I have encountered some difficulties. How big is the scannetv2 dataset used in your paper? I saw 1T on the scannet official website, which may be a terrible number. Looking forward to your reply, thank you!

[llijiang]

Hi, You don't need to download the whole dataset. You could just download [scene_id]_vh_clean_2.ply, [scene_id]_vh_clean_2.labels.ply, [scene_id]_vh_clean_2.0.010000.segs.json and [scene_id].aggregation.json files.

[ghost]

Thank you for your reply, I followed this link to download, https://github.com/ScanNet/ScanNet/tree/master/Tasks/Benchmark, just to download these .ply files, it takes about 500G. How big is the total file size you downloaded?

[Rabbia-Hassan]

@llijiang @yanwei-li @yukang2017 @ghost , hello, did you solve this problem? I am facing the same issue. can you guie me please ? How to transfer the files into train val and text ?

[cChrister]

Thank you for your reply, I followed this link to download, https://github.com/ScanNet/ScanNet/tree/master/Tasks/Benchmark, just to download these .ply files, it takes about 500G. How big is the total file size you downloaded?

I totaly download 20GB, which can run successfully

[cChrister]

@llijiang @yanwei-li @yukang2017 @ghost , hello, did you solve this problem? I am facing the same issue. can you guie me please ? How to transfer the files into train val and text ?

import os
import shutil

def copyfile(srcfile, dstpath):                        # 复制函数
    if not os.path.isfile(srcfile):
        print ("%s not exist!"%(srcfile))
    else:
        fpath,fname=os.path.split(srcfile)             # 分离文件名和路径
        if not os.path.exists(dstpath):
            os.makedirs(dstpath)
        dstpath = os.path.join(dstpath, fname)
        shutil.copy(srcfile, dstpath)                  # 复制文件
        print ("copy %s -> %s"%(srcfile, dstpath))

def mov(t_data, t_files, base_dir):
    for t_file in t_files:
        with open(t_data+'.txt',"r") as f:
            lines = f.readlines()
            for line in lines:
                id = line.strip('\n')
                scrfile = os.path.join(base_dir, id, id + t_file)
                dstpath = os.path.join(target_dir, t_data)
                copyfile(srcfile=scrfile, dstpath=dstpath)

target_dir = '/data0/chenxiang/PointGroup/dataset/scannetv2'

t_data = 'train'
t_files = ['_vh_clean_2.ply', '_vh_clean_2.labels.ply', 
          '_vh_clean_2.0.010000.segs.json', '.aggregation.json']
base_dir = '/data0/chenxiang/data/scans'
mov(t_data=t_data, t_files=t_files, base_dir=base_dir)

t_data = 'val'
t_files = ['_vh_clean_2.ply', '_vh_clean_2.labels.ply', 
          '_vh_clean_2.0.010000.segs.json', '.aggregation.json']
base_dir = '/data0/chenxiang/data/scans'
mov(t_data=t_data, t_files=t_files, base_dir=base_dir)

t_data = 'test'
t_files = ['_vh_clean_2.ply']
base_dir = '/data0/chenxiang/data/scans_test'
mov(t_data=t_data, t_files=t_files, base_dir=base_dir)