lvpengyuan / corner Goto Github PK
View Code? Open in Web Editor NEWMulti-Oriented Scene Text Detection via Corner Localization and Region Segmentation
Multi-Oriented Scene Text Detection via Corner Localization and Region Segmentation
请问如何用自己的数据训练啊?数据集的格式是什么样的啊
Excuse me, The output of the network("out = net(images" in training code) contains four parts(loc_data, conf_data, priors, seg_data). I'm still not sure the meaning of them after reading your paper. Can you help me? Thank you very much!
Traceback (most recent call last): File "eval_all.py", line 380, in <module> main() File "eval_all.py", line 347, in main out, seg_pred, seg_map =net(img) File "/opt/conda/lib/python3.6/site-packages/torch/nn/modules/module.py", line 477, in __call__ result = self.forward(*input, **kwargs) File "/root/fsy_SceneTextRec/Docker-pytorch-0.3/corner-master/dssd.py", line 113, in forward output = self.detect(loc.view(loc.size(0), -1, 4, 4), self.softmax(conf.view(-1, self.num_classes)), self.priors, F.sigmoid(seg_view), F.sigmoid(seg_pred)) File "/root/fsy_SceneTextRec/Docker-pytorch-0.3/corner-master/layers/functions/detection.py", line 50, in forward temp.append(torch.cat((scores[ids[:count]].unsqueeze(1),
I also find that this code can only run in the pytorch environment under pytorch0.4, otherwise we need to change the c code in the roi_pooling module.
I meet this error above when i run on pytorch0.4.1 and run the eval_all.py
I am testing your code. Accordingly,
i) I cloned the repository, executed the .sh
file
ii) Downloaded the model and placed it in weights/
folder
iii) Tried to train the model using the command python train.py
I got the error as follows:
Traceback (most recent call last): File "train.py", line 11, in <module> from data import train_cfg, AnnotationTransform, ICDARDetection, detection_collate, SynthDetection, TD500Detection, MLTDetection File "/home/pg2017/cse/17071016/corner/data/__init__.py", line 1, in <module> from .icdar import ICDARDetection File "/home/pg2017/cse/17071016/corner/data/icdar.py", line 6, in <module> import torchvision.transforms as transforms File "/home/pg2017/cse/17071016/.conda/envs/myenv/lib/python3.7/site-packages/torchvision/__init__.py", line 1, in <module> from torchvision import models File "/home/pg2017/cse/17071016/.conda/envs/myenv/lib/python3.7/site-packages/torchvision/models/__init__.py", line 11, in <module> from . import detection File "/home/pg2017/cse/17071016/.conda/envs/myenv/lib/python3.7/site-packages/torchvision/models/detection/__init__.py", line 1, in <module> from .faster_rcnn import * File "/home/pg2017/cse/17071016/.conda/envs/myenv/lib/python3.7/site-packages/torchvision/models/detection/faster_rcnn.py", line 7, in <module> from torchvision.ops import misc as misc_nn_ops File "/home/pg2017/cse/17071016/.conda/envs/myenv/lib/python3.7/site-packages/torchvision/ops/__init__.py", line 1, in <module> from .boxes import nms, box_iou File "/home/pg2017/cse/17071016/.conda/envs/myenv/lib/python3.7/site-packages/torchvision/ops/boxes.py", line 2, in <module> from torchvision import _C ImportError: /home/pg2017/cse/17071016/.conda/envs/myenv/lib/python3.7/site-packages/torchvision/_C.cpython-37m-x86_64-linux-gnu.so: undefined symbol: _ZN3c1019UndefinedTensorImpl10_singletonE
What is the possible reason for the error? As I have never faced such an error.
Thanks in advance. Waiting for a reply.
File ".\train.py", line 256, in
train()
File ".\train.py", line 241, in train
loss.backward()
File "E:\anaconda\envs\tensorflow\lib\site-packages\torch\tensor.py", line 118, in backward
torch.autograd.backward(self, gradient, retain_graph, create_graph)
File "E:\anaconda\envs\tensorflow\lib\site-packages\torch\autograd_init_.py", line 93, in backward
allow_unreachable=True) # allow_unreachable flag
RuntimeError: element 0 of tensors does not require grad and does not have a grad_fn
from shapely.geometry import box, Polygon isn't exist?
Could anyone tell me the meaning of this code?
Excuse me,how do you process the data of SynthText dataset before training?
安装pytorch0.2.0_3失败,有教程可以参考吗 @lvpengyuan @lvpengyuan
I'm getting the following error:
File "/home/mukut/gitlab/corner/data/icdar.py", line 49, in init
ic13_samples = open(ic13_list_path, 'r').readlines()
IOError: [Errno 2] No such file or directory: '../data/ocr/detection//icdar2013/test_list.txt'
when I executed eval_all.py file.
Any suggestion.
Excuse me, how should I modify the training settings before training to train a new model?
about line 220:
for item in **samples**:
temp.append(Variable(item, volatile=True))
gts.append(temp)
I think, the "samples" should be "sample",thank you.
@lvpengyuan How to achieve SROIE 2019 results
https://rrc.cvc.uab.es/?ch=13&com=evaluation&view=method_info&task=1&m=53374
H&H Lab
Description: we ensemble EAST and multi oriented corner to create a robust scene text detector. To make network learning easier, we modified the mutli-oriented corner network with a new branch borrowed from east added.
@lvpengyuan
Hello, when i ran train.py there was error like below
cannot import name 'getboxes'
There were not any files(getboxes.py, generate_gt) in utils folder
Would you upload both of them?
Thank you in advance
使用自己的数据进行训练,发现速度较AdvancedEast和DB来说都很慢,想问问有没有什么比较好的优化方法
Traceback (most recent call last):
File "/home/work/Code/Corner/train.py", line 255, in
train()
File "/home/work/Code/Corner/train.py", line 236, in train
loss_l, loss_c, loss_s = criterion(out, targets, segs)
File "/home/panda/pytorch_corner/local/lib/python2.7/site-packages/torch/nn/modules/module.py", line 357, in call
result = self.forward(*input, **kwargs)
File "/home/work/Code/Corner/layers/modules/multibox_loss.py", line 79, in forward
loss_c = F.cross_entropy(conf_p, targets_weighted, size_average=False)
File "/home/panda/pytorch_corner/local/lib/python2.7/site-packages/torch/nn/functional.py", line 1161, in cross_entropy
return nll_loss(log_softmax(input, 1), target, weight, size_average, ignore_index, reduce)
File "/home/panda/pytorch_corner/local/lib/python2.7/site-packages/torch/nn/functional.py", line 786, in log_softmax
return torch._C._nn.log_softmax(input, dim)
RuntimeError: dimension specified as 1 but tensor has no dimensions
It seems that ./layers/multibox_loss.py 75 line 'conf_p = conf_data_v[(pos_idx + neg_idx).gt(0)].view(-1, self.num_classes)' have bug?
Hey @lvpengyuan I really appreciate your work of the paper 'Multi-Oriented Scene Text Detection using corner localisation and region segmentation'.
However, I am facing difficulty in acquiring the MLT dataset that you have used in your implementation. Can you please provide me the dataset?
Hi,
Since PyTorch 1 torch.utils.ffi is deprecated. Has anyone tried to rewrite it to current C++ API?
Regards,
在哪里下载预训练的模型?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.