-
Notifications
You must be signed in to change notification settings - Fork 8k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
A Simple and Efficient Network for Small Target Detection #4213
Comments
I don't see [yolo] layer in you cfg-file. |
The cfg-file: network.cfg.txt
The proposed network in the paper does not have any [yolo] or [cost] layers. Based on the yolov3-tiny.cfg file, I changed the activation function of last layer to linear and added a [yolo] layer after it (network_with_yolo.cfg.txt). Now it can be trained but the performance is weaker than YoloV3-Tiny. No NaN for |
I thought that in the table, left column is the architecture of authors proposed network and right column is the architecture of Tiny YoloV3 and each column presents a separate independent architecture. Therefore, the [yolo] layer you mentioned, is in the Tiny YoloV3 not the proposed network. |
Yes, sure, you are right ) But still no detection network can work without a detection head: [yolo], SSD, Faster RCNN, ... |
Thanks. So there may be a mistake in the table. As I mentioned before, adding a [yolo] layer after the last convolution layer did not give any interesting results:
Despite the [yolo] layer, is the configuration in network_with_yolo.cfg.txt conforming with the proposed network in the paper? I used [route] layer for Concatenation layers and [reorg3d] layer for the Passthrough layer. |
Yes, it seems network_with_yolo.cfg.txt conforming with the proposed network in the paper
Thats right. Try to use in the [yolo] layer
|
A custom dataset. I have not tested the datasets used in the paper.
1734 images.
About 30x30.
chart.png for yolov3-tiny.cfg.txt: chart.png for network_with_yolo.cfg.txt: Note that:
Without these changes the mAP was lower with We have a separate test set. Here are the results of With best weights using yolov3-tiny.cfg.txt:
With best weights using network_with_yolo.cfg.txt:
|
Are mAPs on the charts for Training or Validation dataset? |
Validation |
Why on the chart you get 99.9% but for ./darknet detector map ... you get 67.20% for network_with_yolo.cfg.txt ? |
I used a separate test set for |
Did you get Training/Valid/Test dataset by randomly uniform dividing single dataset to 80%/10%/10%? |
Train and valid sets are selected randomly from a single dataset with 1734 images for train and 530 images for valid . But the test set is an independent set. |
So may be this is the reason. Your train for one objects, but test for others. |
Yes, you are right |
@mrhosseini Hi, When I using network_with_yolo.cfg I’m faced with this error. cuDNN status Error in: file: ....\src\convolutional_layer.c : get_workspace_size16() I have 18 classes and I just changed: |
@zpmmehrdad |
@mrhosseini Hi, Thanks, What CUDNN and CUDA version are you using? |
@zpmmehrdad
|
@AlexeyAB Hi, I'm using output: compute_capability = 610, cudnn_half = 0 cuDNN Error: CUDNN_STATUS_NOT_SUPPORTED: No error |
@zpmmehrdad What GPU do you use? |
@AlexeyAB Hi, |
@AlexeyAB Hi, |
@mrhosseini Hello? I'm also studying this field recently. Are you running on windows? If so, can you send me a copy of your compiled Darknet and pack it for me? I encountered a lot of errors in compiling. My email is 1373890292@qq.com,I look forward to your reply. |
Hi @leiyaohui , unfortunately I use Ubuntu. Try one of the methods here. You may open a new issue if encountered with errors. |
Did you write the expansion convolution or did it come with Darknet itself?
---原始邮件---
发件人:"mrhosseini"<notifications@github.com>;
发送时间:2019年12月9日(星期一) 下午5:57
收件人:"AlexeyAB/darknet"<darknet@noreply.github.com>;
抄送人:"leiyaohui"<1373890292@qq.com>;"Mention"<mention@noreply.github.com>;
主题:Re: [AlexeyAB/darknet] A Simple and Efficient Network for SmallTarget Detection (#4213)
Hi @leiyaohui , unfortunately I use Ubuntu. Try one of the methods here. You may open a new issue if encountered with errors.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub, or unsubscribe.
|
The dilated convolution is implemented in this repository. You can use this configuration file for the proposed network of the paper which mentioned above. |
Hi,
This paper proposes a new network configuration for small target detection and claims that it has a performance near YoloV3 while a speed near YoloV3-Tiny. The main idea is to use dilated and 1x1 convolutions.
I tried to implement the network using this repo but in training always get NaN for loss and avg loss.
Here is the configuration that I used for single class detection:
and this is the proposed network in the paper:
Any advice for solving the problem?
The text was updated successfully, but these errors were encountered: