DocF/multispectral-object-detection

Issues

• 关于CFT weight的使用

  #62 opened a year ago by nlynzb
  8

• How to get train.txt

  #80 opened 4 months ago by mengyin369
  4

• VEDAI数据集划分

  #36 opened 2 years ago by liufangcen
  5

• FLIR实验参数

  #57 opened a year ago by duoxiangqinzuo
  8

• 关于找不到FFVIP的缓存图片

  #31 opened 3 years ago by cenchaojun
  2

• MR评价指标

  #64 opened a year ago by jianguo520
  1

• 请问下LLVIP是怎么转换YOLO格式的，infrared和visible分别操作？

  #38 opened 2 years ago by XuKer
  1

• 刚入门的小白有些问题想要请教

  #52 opened 2 years ago by dududududukehan
  1

• 抱歉刚入门复现此项目，请问转出来的.txt格式的标签文件的位置应该放在哪里

  #40 opened 2 years ago by jayendsff
  16

• 关于flir数据集推理类别，我训练的时候是三个类别，但将生成权重应用于推理就变成了只有person一类，所有的车辆都错检为person

  #34 opened 3 years ago by HUANG7M
  4

• 数据集划分在哪里？

  #69 opened 10 months ago by dyx30
  3

• 在M3FD数据集上训练结果不如yolov5-5.0的训练结果

  #79 opened 6 months ago by mydy2024
  0

• 关于ADD2模型，为什么在add命令上会有不同，难道不是相同位置元素相加吗？请问为什么这样设置add

  #75 opened 9 months ago by leileilei2000
  2

• 关于VEDAI数据集训练问题

  #68 opened a year ago by eqwrqwrqr
  2

• 关于测试结果

  #43 opened 2 years ago by medicineqian
  4

• 红外图像不应该是单通道吗？为什么图中是640*640*3？

  #48 opened 2 years ago by CQUMASTER
  3

• test.py可以正常运行但是没有任何图像输出，请问要怎么修改

  #78 opened 9 months ago by leileilei2000
  0

• 请问您的代码是基于yolov5的哪个版本改了，这样能方便我跑下单模态对比实验，谢谢

  #76 opened 9 months ago by mujianyu
  1

• bug：嵌入位置的参数self.pos_emb不更新

  #77 opened 9 months ago by Yipzcc
  0

• VEDAI是只训练了fold1吗？

  #74 opened 9 months ago by leileilei2000
  0

• Input shape of Vedai

  #72 opened 9 months ago by love-whut
  1

• FLIR Google Drive Link points to random eBay listing

  #63 opened a year ago by ntinou
  8

• 请问如何将CFT移植到Faster rcnn中对特征进融合？

  #73 opened 9 months ago by chenyao1216
  0

• 作者您好，我对您的工作很感兴趣。在项目中并没有找到transformer或者CFT相关的代码，请问CFT是如何实现的~

  #71 opened 9 months ago by chenyao1216
  0

• 您好，请问我的数据集是640*480的，是否需要处理成640*640大小，另外，彩色的红外图像需要处理成灰度红外图像吗？

  #61 opened a year ago by shensongli
  1

• 有没有人知道这个项目yolov5框架的版本啊

  #67 opened a year ago by 225ceV
  2

• 测试

  #70 opened 10 months ago by Highwindsfindtheirwayhome
  1

• FLIR上复现结果比论文高10%

  #58 opened a year ago by shensongli
  11

• 训练和测试数据加载尺度不一致的问题

  #45 opened 2 years ago by ysh-strive
  1

• 运行test.py为啥没有绘制PR、P、R等结果图

  #66 opened a year ago by seenthestars
  0

• How to accelerate the model with tensorrt?

  #65 opened a year ago by Archie945
  0

• 关于在LLVIP数据集上的复现结果

  #44 opened 2 years ago by XueZ-phd
  7

• 请问最后得到的loss曲线中的val box loss为什么是上升趋势呢

  #59 opened a year ago by shensongli
  19

• 参数量和浮点计算量的计算方法

  #60 opened a year ago by duoxiangqinzuo
  0

• About Conv

  #56 opened a year ago by myc1998
  0

• Annotation

  #55 opened 2 years ago by Inovative-UMS
  0

• RuntimeError: result type Float can't be cast to the desired output type long int

  #47 opened 2 years ago by minkyukang707
  6

• 请问各位，align_train.txt、align_validation.txt这样类似的四个文件应该怎么生成呀

  #53 opened 2 years ago by dududududukehan
  1

• FLIR dataset annotation

  #54 opened 2 years ago by nineRishav
  0

• 您好，请问如何计算Log Average Miss Rate？我在库里没看到相应的代码，谢谢

  #32 opened 3 years ago by mzzjuve
  1

• SSD on the paper

  #51 opened 2 years ago by heitorrapela
  0

• 测试结果map过高

  #50 opened 2 years ago by zx0v0
  1

• 如何将all annotations 转化成YOLOv5 format？

  #42 opened 2 years ago by aiLab-wu
  1

• TypeError: 'NoneType' object is not subscriptable

  #49 opened 2 years ago by kyloyang1
  0

• Error in the Two Stream Detection Code

  #46 opened 2 years ago by anilcanbulut
  1

• Input shape of LLVIP YOLOV5L

  #33 opened 3 years ago by XiongZhongxia
  3

• Issue in path

  #41 opened 2 years ago by nineRishav
  1

• 实现双流模态的代码

  #39 opened 2 years ago by chenqq1
  0

• 重训练不出作者给的结果

  #37 opened 2 years ago by GXQ2020
  0

• 'SPPF'

  #35 opened 3 years ago by 1phil1
  1