yimiandai/open-acm

code and trained models for "Asymmetric Contextual Modulation for Infrared Small Target Detection"

Hi there, I'm Yi-Mian Dai (戴一冕 in Chinese)! 👋

I am a PhD student at Nanjing University of Aeronautics & Astronautics (NUAA).

🔭 Reseach-wise, I mainly focus on:

• Small Object Detection
• Neural Architecture Design

😄 I am open to:

• collaboration opportunities (anytime & anywhere & any type)
• postdoc position, research internships, or visiting scholar

📫 How to reach me:

• Email: yimian [dot] dai [at] gmail.com
• WeChat (微信) (Anoymous requests are not welcome)

Publications:

• Yimian Dai, Yiquan Wu, Fei Zhou, Kobus Barnard. Attentional Local Contrast Networks for Infrared Small Target Detection[J]. Accepted by IEEE Transactions on Geoscience and Remote Sensing. (code)
• Yimian Dai, Yiquan Wu, Fei Zhou, Kobus Barnard. Asymmetric Contextual Modulation for Infrared Small Target Detection[C]. Accepted by IEEE Winter Conference on Applications of Computer Vision, WACV 2021. (dataset, code)
• Yimian Dai, Fabian Gieseke, Stefan Oehmcke, Yiquan Wu, Kobus Barnard. Attentional Feature Fusion[C]. Accepted by IEEE Winter Conference on Applications of Computer Vision, WACV 2021. (code)
• Yimian Dai, Stefan Oehmcke, Fabian Gieseke, Yiquan Wu, Kobus Barnard. Attention as Activation[C]. Accepted by 25th International Conference on Pattern Recognition (ICPR 2020), Oral. (code)
• Yimian Dai, Yiquan Wu. Reweighted Infrared Patch-Tensor Model with Both Nonlocal and Local Priors for Single-Frame Small Target Detection[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2017, 10(8), 3752-3767. (code)

Software:

• iNSFC: An awesome LaTeX template for NSFC proposal
• NUAAletter: A LaTeX Letter template for NUAA

Services:

• Journal Reviewer
  • IEEE Transactions on Geoscience and Remote Sensing (TGRS)
  • IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (JSTARS)
  • IEEE Geoscience and Remote Sensing Letters (GRSL)
  • Infrared Physics & Technology

Miscellaneous:

• PhD Thesis (in Chinese)

open-acm's Issues

代码问题

您好，我运行代码的时候发现只有training loss的结果，但IoU和nIoU都为0，请问，是我代码有的地方没有改吗？

关于传统方法（例如NIPPS和RIPT）的IoU和nIoU的计算

戴博士您好，很抱歉再次打扰到您，我还有一个问题想向您请教下。
我们在进行对比实验时，传统方法NIPPS和RIPT等，如何能够得到IoU和nIoU这类指标呢？

期待作者的pytorch版本的代码

非常感谢作者的开源代码，您在这方面取得的成就是令人钦佩的，同时，也非常感谢你开源了相关的代码。但是由于我自身学习方面的不足，目前仅仅比较熟悉pytorch相关的内容，希望您能够尽快上传pytorch版本的已经开源的版本的论文及代码！！！
非常期待您的更新。

How to retrain open-acm on SIRST dataset ?

I would like to re-train your model on the SIRST dataset. Are the model parameters and the logs provided is obtained by training the model using SIRST dataset or a different dataset ? Also, I am not sure about the values for the variables dataset, train-split, and val-split. I am little confused as the default values for dataset are DENTIST, Iceberg, StopSign. Could you please help me on the parameter values ? Thanks in advance.

论文的疑问

您好，请问你ACM论文中使用的sirst数据集图片在训练和测试被缩放到什么大小？

您好，能够提供一下，测试结果的可视化脚本？？？

The location of batchnorm2D

Why did you locate the batchnorm2D behind the input image (denoted as x in the network; hybridsquential)

Any advantage is there?

论文的疑问

您好，请问论文中您在sirst427数据集上的实验是怎么划分的呢？
是训练集256张，验证集85张，测试集86张吗？

mxnet报错

OSError: libnccl.so.2: cannot open shared object file: No such file or directory
换了几个版本，一直报这个错，各位大侠有遇到吗，怎么解决的？

mxnet环境问题

我在使用该代码时，首先按照readme中的步骤配置mxnet环境，详细操作为：pip install --upgrade mxnet-cu100 gluoncv，
相关版本为 mxnet：1.5.0，gluoncv：0.8.0
但在跑实验时遇到以下问题：
E:\Anaconda3\envs\zmj\python.exe "C:/Users/hp/Desktop/deep learning/open-acm-master/train_dentist.py"
Traceback (most recent call last):
File "C:/Users/hp/Desktop/deep learning/open-acm-master/train_dentist.py", line 14, in
from gluoncv.utils import makedirs, LRScheduler, LRSequential
File "E:\Anaconda3\envs\zmj\lib\site-packages\gluoncv_init_.py", line 12, in
from . import data
File "E:\Anaconda3\envs\zmj\lib\site-packages\gluoncv\data_init_.py", line 31, in
from .kitti.kitti_dataset import KITTIRAWDataset, KITTIOdomDataset
File "E:\Anaconda3\envs\zmj\lib\site-packages\gluoncv\data\kitti_init_.py", line 2, in
from .kitti_dataset import *
File "E:\Anaconda3\envs\zmj\lib\site-packages\gluoncv\data\kitti\kitti_dataset.py", line 21, in
from .kitti_utils import generate_depth_map
File "E:\Anaconda3\envs\zmj\lib\site-packages\gluoncv\data\kitti\kitti_utils.py", line 12, in
from mxnet.util import is_np_array
ImportError: cannot import name 'is_np_array' from 'mxnet.util' (E:\Anaconda3\envs\zmj\lib\site-packages\mxnet\util.py)

Process finished with exit code 1

请问作者现在所用到的mxnet和gluoncv的版本是什么呢？谢谢作者对我的问题进行解答，不胜感激！

测试数据问题请教

作者您好，最近在准备一篇论文，想要与您所提出的这个方法进行对比。由于刚开始接受深度学习的红外小目标检测，所以对相关指示与框架运用的都不够熟练，因此想要请教您以下问题：

1. 该网络能够用您训练好的数据直接进行我手头数据集的测试呢，测试的部分即为mode='testval'这部分，不知道我理解是否正确；

2. 对数据集是否要进行预处理以满足此框架的需求，我看有一部分输入要求是.txt文件，而非.bmp和.png之类的图像文件；

3. 在此框架下测试其他数据集时，我该具体调整哪部分代码呢？

由于是深度学习小白，所以希望大佬能将以上问题进行回复，我很很感激您哒～

祝您论文高产且全是高水平论文！

关于测试问题

你好，您给出的训练代码已经跑起来了，但是只给出了IOU的结果，请问怎么在测试数据上进行可视化的显示呢？直观的看到检测结果

How to satisfy MXNet requirement for open-acm?

I'm using Jetson AGX Orin with Jetpack 5.1.2 and Cuda 11.4 and I'm unable to satisfy the MXNet requirement and run open-acm. MXNet is outdated and not maintained, thus building it with recent technology is not feasible. I've tried different branches of MXNet such as '1.x', '1.9.x' and '1.9.1' but no luck (can't configure for c++14). Has anyone encountered such an issue regarding the installation of open-acm. If anyone has successfully installed open-acm on this platform I'd like to know how.

IoU和nIoU阈值统一问题

请问IoU和nIoU中用于过滤的阈值是否应该一致？

兄弟能不能加一下您的微信

关于代码中IoU和nIoU的计算问题

非常感谢您在红外小目标检测中的工作，我在仔细阅读您的代码后有一点小疑问，想向您请教。
在IoU和mIoU的指标计算中：
_, IoU = self.iou_metric.get()
_, nIoU = self.nIoU_metric.get()
return的值都是mIoU=IoU.mean(),请问一下这二者有什么区别？期望回复，谢谢

单帧处理时间

作者您好，我想问一下这个代码处理一帧需要的时间大概是多久呢