@@ -17,7 +17,8 @@ docker pull beniz/deepdetect_cpu
17
17
docker run -d -p 8080:8080 beniz/deepdetect_cpu
18
18
```
19
19
20
- Right now, the only supported installation of DeepDetect that works with DeepSort is the deepdetect_cpu container.
20
+ Right now, the only supported installation of DeepDetect that works with DeepSort is the deepdetect_cpu container,
21
+ because it contain the good path for the pre-installed ` resnet-50 ` and ` googlenet ` models.
21
22
22
23
Then, download the latest DeepSort release from https://github.com/CorentinB/DeepSort/releases
23
24
@@ -37,9 +38,9 @@ For more informations, refeer to the helper:
37
38
./DeepSort --help
38
39
39
40
[-u|--url] is required
40
- usage: DeepSort [-h|--help] -u|--url "<value>" -i|--input "<value>"
41
- [-n|--network (resnet-50|googlenet)] [-R|--recursive ]
42
- [-j|--jobs <integer>] [-d|--dry-run]
41
+ usage: deepsort [-h|--help] -u|--url "<value>" -i|--input "<value>"
42
+ [-o|--output "<value>"] [- n|--network (resnet-50|googlenet)]
43
+ [-R|--recursive] [- j|--jobs <integer>] [-d|--dry-run]
43
44
44
45
AI powered image tagger backed by DeepDetect
45
46
@@ -48,6 +49,9 @@ Arguments:
48
49
-h --help Print help information
49
50
-u --url URL of your DeepDetect instance (i.e: http://localhost:8080)
50
51
-i --input Your input folder.
52
+ -o --output Your output folder, if output is set, original files will
53
+ not be renamed, but the renamed version will be copied in
54
+ the output folder.
51
55
-n --network The pre-trained deep neural network you want to use, can be
52
56
resnet-50 or googlenet. Default: resnet-50
53
57
-R --recursive Process files recursively.
@@ -59,7 +63,7 @@ Arguments:
59
63
60
64
- [X] Getting docker out of the loop (each user install his own DeepDetect)
61
65
- [X] ResNet 50 integration
62
- - [ ] Output folder (copy and not rename)
66
+ - [X ] Output folder (copy and not rename)
63
67
- [ ] NSFW tagging (Yahoo open_nsfw)
64
68
- [ ] XMP metadata writing
65
69
- [ ] GPU support
0 commit comments