Anzhc commited on
Commit
4e98433
·
verified ·
1 Parent(s): 3ebf5bd

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +12 -12
README.md CHANGED
@@ -13,14 +13,14 @@ library_name: ultralytics
13
  ### Anzhc's Face segmentation:
14
  #### Universal:
15
  Series of models aiming at detecting and segmenting face accurately. Trained on closed dataset i annotated myself.
16
- | Model | Target | mAP 50 | mAP 50-95 |Classes |Training Resolution|
17
- | --------------------------- | --------------------- | ----------------------------- | ------------------------- |---------------|-------------------|
18
- | Anzhc Face -seg.pt | Face: illustration, real | LOST DATA | LOST DATA |2(male, female)| 640|
19
- | Anzhc Face -seg-hd.pt | Face: illustration, real | LOST DATA | LOST DATA |2(male, female)|1024|
20
- | Anzhc Face seg 640 v2 y8n.pt | Face: illustration, real | 0.872(box),0.872(mask) | 0.835(box),0.752(mask)|1(face) | 640|
21
- | Anzhc Face seg 768 v2 y8n.pt | Face: illustration, real | 0.86(box),0.86(mask) | 0.81(box),0.726(mask) |1(face) | 768|
22
- | Anzhc Face seg 768MS v2 y8n.pt | Face: illustration, real | 0.866(box),0.866(mask) | 0.816(box),0.72(mask) |1(face) | 768|(Multi-scale)|
23
- | Anzhc Face seg 1024 v2 y8n.pt | Face: illustration, real | 0.872(box),0.872(mask) | 0.804(box),0.726(mask)|1(face) | 1024|
24
 
25
  Take those stats with a grain of salt, since im pretty sure i re-scrambled dataset partition after training those models ages ago.
26
  Benchmark was performed in 640px.
@@ -32,10 +32,10 @@ Difference in v2 models are only in their target resolution, so their performanc
32
  #### Real Face, gendered:
33
  Trained only on real photos for the most part, so will perform poorly with illustrations, but is gendered, and can be used for male/female detection stack.
34
 
35
- | Model | Target | mAP 50 | mAP 50-95 |Classes |Training Resolution|
36
- | --------------------------- | --------------------- | ----------------------------- | ------------------------- |---------------|-------------------|
37
- | Anzhcs ManFace v02 1024 y8n.pt | Face: real | 0.883(box),0.883(mask) | 0.778(box), 0.704(mask) |1(face)|1024|
38
- | Anzhcs WomanFace v05 1024 y8n.pt | Face: real | 0.82(box),0.82(mask) | 0.713(box), 0.659(mask) |1(face)|1024|
39
 
40
  Benchmark was performed in 640px.
41
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/633b43d29fe04b13f46c8988/W0vhyDYLaXuQnbA1Som8f.png)
 
13
  ### Anzhc's Face segmentation:
14
  #### Universal:
15
  Series of models aiming at detecting and segmenting face accurately. Trained on closed dataset i annotated myself.
16
+ | Model | Target | mAP 50 | mAP 50-95 |Classes |Dataset size|Training Resolution|
17
+ | --------------------------- | --------------------- | ----------------------------- | ------------------------- |---------------|------------|-------------------|
18
+ | Anzhc Face -seg.pt | Face: illustration, real | LOST DATA | LOST DATA |2(male, female)|LOST DATA| 640|
19
+ | Anzhc Face -seg-hd.pt | Face: illustration, real | LOST DATA | LOST DATA |2(male, female)|LOST DATA|1024|
20
+ | Anzhc Face seg 640 v2 y8n.pt | Face: illustration, real | 0.872(box),0.872(mask) | 0.835(box),0.752(mask)|1(face) |~500| 640|
21
+ | Anzhc Face seg 768 v2 y8n.pt | Face: illustration, real | 0.86(box),0.86(mask) | 0.81(box),0.726(mask) |1(face) |~500| 768|
22
+ | Anzhc Face seg 768MS v2 y8n.pt | Face: illustration, real | 0.866(box),0.866(mask) | 0.816(box),0.72(mask) |1(face) |~500| 768|(Multi-scale)|
23
+ | Anzhc Face seg 1024 v2 y8n.pt | Face: illustration, real | 0.872(box),0.872(mask) | 0.804(box),0.726(mask)|1(face) |~500| 1024|
24
 
25
  Take those stats with a grain of salt, since im pretty sure i re-scrambled dataset partition after training those models ages ago.
26
  Benchmark was performed in 640px.
 
32
  #### Real Face, gendered:
33
  Trained only on real photos for the most part, so will perform poorly with illustrations, but is gendered, and can be used for male/female detection stack.
34
 
35
+ | Model | Target | mAP 50 | mAP 50-95 |Classes |Dataset size|Training Resolution|
36
+ | --------------------------- | --------------------- | ----------------------------- | ------------------------- |---------------|------------|-------------------|
37
+ | Anzhcs ManFace v02 1024 y8n.pt | Face: real | 0.883(box),0.883(mask) | 0.778(box), 0.704(mask) |1(face)|~340|1024|
38
+ | Anzhcs WomanFace v05 1024 y8n.pt | Face: real | 0.82(box),0.82(mask) | 0.713(box), 0.659(mask) |1(face) |~600|1024|
39
 
40
  Benchmark was performed in 640px.
41
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/633b43d29fe04b13f46c8988/W0vhyDYLaXuQnbA1Som8f.png)