biubug6 Pytorch_Retinaface issues

biubug6 / Pytorch_Retinaface

Retinaface get 80.99% in widerface hard val using mobilenet0.25.

MIT License

2.63k stars 774 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

How to use negative data for training?

#111 RocketWill opened 4 years ago
1
SSH code is different from paper

#110 gaoruiyang closed 4 years ago
1
same parameter with Mxnet

#109 Liuqiyuan670 opened 4 years ago
2
使用该工程检测有误检，请问应该怎么调节？(阈值之类的参数已经调过，没有作用)

#108 jingruhou opened 4 years ago
7
Identifying small faces in large images

#107 ashlinghosh opened 4 years ago
3
what does the performace mean?

#106 xman1991 opened 4 years ago
0
512-D Feature Embedding

#105 MozhganPourKeshavarz opened 4 years ago
1
mobilenet-0.25在i7 cpu前传时间60ms左右

#104 GengCauWong opened 4 years ago
1
How can i training Resnet50 model use single GPU 8G?

#103 Yooben-Li opened 4 years ago
2
cpu consume

#102 wanjinchang opened 4 years ago
0
Is your input not be normalized? Is that Ok?

#101 zhangluustb closed 4 years ago
0
train.py输出时，会多次打印cfg和网络结构，这是什么原因啊

#100 Tallisgo opened 4 years ago
3
读取文件夹下不同尺寸的图片，前向速度非常慢

#99 xiaocmxiao opened 4 years ago
0
Question about onnx

#98 zdaiot opened 4 years ago
0
Worse performance on windows than Ubuntu

#97 Arki99 closed 4 years ago
1
怎么更改anchor的比例

#96 ghost opened 4 years ago
2
data result on FDDB

#95 dwisnantoputro opened 4 years ago
0
Annotations zip file doesn't contain wider_val.txt

#94 yossibiton opened 4 years ago
1
Face alignment

#93 dgy20122016 opened 4 years ago
3
I have a question about the process of WIDERFACE dataset

#92 yBeOne opened 4 years ago
1
hell. i have a question about encode and decode

#91 DAEHEESHIN opened 4 years ago
0
Model does not converge when i change the input size from 640*640 to 360*640.

#90 ThomsonW opened 4 years ago
0
retinaface (resnet50) in tensorrt

#89 wang-xinyu closed 4 years ago
1
Invalid images in the dataset?

#88 sri9s opened 4 years ago
2
关于在线样本挖掘

#87 lycwgf opened 4 years ago
2
No Dense Regression Loss?

#86 jjjjohnson opened 4 years ago
1
Hi, I found some thing strange, can you help me better understand these?

#85 young169 opened 4 years ago
0
关于PriorBox

#84 anguoyang opened 4 years ago
0
The output bboxes is so big??

#83 szupzp opened 4 years ago
3
使用自制数据集进行训练时遇到问题

#82 shawnco411 closed 4 years ago
0
Why not use context modelling proposed on the paper?

#81 BlueCCircles closed 4 years ago
9
face min size

#80 alicera opened 4 years ago
0
inference time is longer than 5-6ms produced by author.And different size input, but the inference time is same.

#79 Govan111 opened 4 years ago
2
_det.txt file missing

#78 anguoyang closed 4 years ago
0
the Inference time is too long

#77 liguoyu666 closed 4 years ago
2
GPU on test?

#76 Alpang54 closed 4 years ago
0
What is a resume net? Can someone please elaborate?

#75 neelabh17 closed 3 years ago
3
PriorBox use much time

#74 zhouyongxiu closed 4 years ago
6
Square face bounding box

#73 miracleyoo opened 4 years ago
1
Any difference between this mnet.25 model and that of Face-Detector-1MB-with-landmark?

#72 ufoym closed 4 years ago
0
ValueError: 'box_overlaps.pyx' doesn't match any files

#71 Coderx7 closed 4 years ago
1
请问，在将检测到的人脸框输入到人脸识别模块之前，是不是首先要做一次人脸对齐映射？

#70 klq-Y opened 4 years ago
1
train with efficentnet backbone

#69 mehrazi opened 4 years ago
0
Training model without landmarks

#68 vovaxnz opened 4 years ago
10
R50 train parameter

#67 dianxin556 opened 4 years ago
4
Return IntermediateLayers

#66 FrankXinqi opened 4 years ago
0
why the network doesn't converage? I use the original wild face dataset

#65 121649982 opened 4 years ago
1
training strategy

#64 goodgoodstudy92 opened 4 years ago
1
Any one replace the FPN with ASFF or BiFPN?

#63 hegc opened 4 years ago
8
FPS Problem

#62 DingtianX opened 4 years ago
6

Previous Next