DBNet.pytorch icon indicating copy to clipboard operation
DBNet.pytorch copied to clipboard

A pytorch re-implementation of Real-time Scene Text Detection with Differentiable Binarization

Results 103 DBNet.pytorch issues
Sort by recently updated
recently updated
newest added

Traceback (most recent call last): File "tools/train.py", line 79, in main(config) File "tools/train.py", line 37, in main train_loader = get_dataloader(config['dataset']['train'], config['distributed']) File "/home/bowen/下载/DBNet.pytorch-master/data_loader/__init__.py", line 84, in get_dataloader _dataset = get_dataset(data_path=data_path,...

An error occurred during training, such as "assert os.path.exists(args.config_file) AssertionError" How to solve it?

您好,我想问一下,当前的网络结构,可以怎么改造一下,实现印刷文本和手写文本得到各自类别的mask吗,就跟语义分割一样。

请问windows下能运行吗? 我训练报错: ![image](https://user-images.githubusercontent.com/40531744/103343238-9b097380-4ac6-11eb-85c1-36a68ff58d7f.png) batch_size和图片大小都设置得很小了,仍然这样; 运行predict.py也报错: (py37) E:\pro\DBNet.pytorch-master> 环境为win10,pytorch为1.6.0

作者您好,试用了下您开源的DBnet,在训练icdar2015时,使用resnet18 始终得不到80.6的hmean,recall只有0.5,precision 0.75,F1为0.60

在源码中,在base_dataset.py中的load_data()函数没有返回值,导致了data_list是NONE,正常返回的data_list是如何的呢,之前load_data()中是否是有代码的呢??

def parse_config(config: dict) -> dict: import anyconfig base_file_list = config.pop('base') base_config = {} for base_file in base_file_list: tmp_config = anyconfig.load(open(base_file, 'rb')) if 'base' in tmp_config: tmp_config = parse_config(tmp_config) # anyconfig.merge(tmp_config,...

order_points_clockwise,某些倾斜角度较大时候,会导致错误,例子:45.45,226.83,11.87,181.79,183.84,13.1,233.79,49.95,

![微信图片_20200911224836](https://user-images.githubusercontent.com/11660784/93001167-b4aa9d80-f55f-11ea-9ecb-18c27f725879.png) 训练loss降到了0.1,precision,recall和hmean都是0.99,但是预测结果框会偏小,尤其是下边界和右边界会明显的压着文字,每一张都是这样,请问有碰到过这个问题吗,可能的原因是什么,可以解决吗?