Hugging Face's logo

Datasets:
issai
/
TFW

Tasks:
Object Detection
Modalities:
Image
Size:
1K<n<10K
Tags:
thermal faces
thermal facial landmarks
License:
Dataset card Data Studio Files Community
akuzdeuov commited on
Commit
9ce88f0
·
verified ·
1 Parent(s): b990787

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +23 -52
README.md CHANGED
@@ -1,49 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
1
  # TFW: Annotated Thermal Faces in the Wild Dataset Card
 
2
 
3
- **Summary:**
4
-
5
- The TFW dataset provides a collection of thermal face images captured in diverse environments: controlled indoor (c-indoor), semi-controlled indoor (s-indoor), and uncontrolled outdoor (u-outdoor). It's annotated with bounding boxes and 5-point facial landmarks. The dataset is accompanied by pre-trained YOLOv5 and YOLO5Face models for thermal face detection, trained and evaluated on the TFW data. The dataset builds upon the SpeakingFaces dataset for its c-indoor portion.
6
-
7
- **Dataset Statistics:**
8
-
9
- | Environment | Subjects | Images | Labeled Faces | Visual Pair |
10
- |--------------|----------|--------|----------------|-------------|
11
- | c-indoor | 142 | 5,112 | 5,112 | yes |
12
- | s-indoor | 9 | 780 | 1,748 | yes |
13
- | u-outdoor | 15 | 4,090 | 9,649 | no |
14
- | **combined** | **147** | **9,982** | **16,509** | yes & no |
15
-
16
-
17
- **Example Annotated Images:**
18
 
 
 
 
 
 
 
 
 
 
19
 
20
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/67056b2e6409e548690b1b6f/uWWvF2xxBfy8jNxaP2Kwk.png)
21
 
22
 
23
- **Pre-trained Models:**
24
-
25
- | Model | Backbone | c-indoor AP<sub>50</sub> | u-outdoor AP<sub>50</sub> | Speed (ms) V100 b1 | Params (M) | Flops (G) @512x384 |
26
- |-----------------|--------------|-------------------------|-------------------------|----------------------|-------------|--------------------|
27
- | YOLOv5n | CSPNet | 100 | 97.29 | 6.16 | 1.76 | 0.99 |
28
- | YOLOv5n6 | CSPNet | 100 | 95.79 | 8.18 | 3.09 | 1.02 |
29
- | YOLOv5s | CSPNet | 100 | 96.82 | 7.20 | 7.05 | 3.91 |
30
- | YOLOv5s6 | CSPNet | 100 | 96.83 | 9.05 | 12.31 | 3.88 |
31
- | YOLOv5m | CSPNet | 100 | 97.16 | 9.59 | 21.04 | 12.07 |
32
- | YOLOv5m6 | CSPNet | 100 | 97.10 | 12.11 | 35.25 | 11.76 |
33
- | YOLOv5l | CSPNet | 100 | 96.68 | 12.39 | 46.60 | 27.38 |
34
- | YOLOv5l6 | CSPNet | 100 | 96.29 | 15.73 | 76.16 | 110.2 |
35
- | YOLOv5n-Face | ShuffleNetv2 | 100 | 95.93 | 10.12 | 1.72 | 1.36 |
36
- | YOLOv5n6-Face | ShuffleNetv2 | 100 | 95.59 | 13.30 | 2.54 | 1.38 |
37
- | YOLOv5s-Face | CSPNet | 100 | 96.73 | 8.29 | 7.06 | 3.67 |
38
- | YOLOv5s6-Face | CSPNet | 100 | 96.36 | 10.86 | 12.37 | 3.75 |
39
- | YOLOv5m-Face | CSPNet | 100 | 95.32 | 11.01 | 21.04 | 11.58 |
40
- | YOLOv5m6-Face | CSPNet | 100 | 96.32 | 13.97 | 35.45 | 11.84 |
41
- | YOLOv5l-Face | CSPNet | 100 | 96.18 | 13.57 | 46.59 | 25.59 |
42
- | YOLOv5l6-Face | CSPNet | 100 | 95.76 | 17.29 | 76.67 | 113.2 |
43
-
44
-
45
  **Citation:**
46
-
47
  ```bibtex
48
  @ARTICLE{9781417,
49
  author={Kuzdeuov, Askat and Aubakirova, Dana and Koishigarina, Darina and Varol, Huseyin Atakan},
@@ -54,17 +38,4 @@ The TFW dataset provides a collection of thermal face images captured in diverse
54
  number={},
55
  pages={2084-2094},
56
  doi={10.1109/TIFS.2022.3177949}}
57
- ```
58
-
59
- **GitHub Repository:** [https://github.com/IS2AI/TFW](https://github.com/IS2AI/TFW)
60
-
61
-
62
- **Paper:** [https://www.techrxiv.org/articles/preprint/TFW_Annotated_Thermal_Faces_in_the_Wild_Dataset/17004538](https://www.techrxiv.org/articles/preprint/TFW_Annotated_Thermal_Faces_in_the_Wild_Dataset/17004538) (Open Access) [https://ieeexplore.ieee.org/abstract/document/9781417](https://ieeexplore.ieee.org/abstract/document/9781417) (IEEE)
63
-
64
- **YouTube Video:** [https://www.youtube.com/watch?v=QgXi3rLv1jM](https://www.youtube.com/watch?v=QgXi3rLv1jM)
65
-
66
- <!-- [Insert demo GIF here from the provided link: https://github.com/IS2AI/TFW/blob/main/figures/demo.gif]
67
-
68
- [Insert YOLOv5 results image here from the provided link: https://github.com/IS2AI/TFW/blob/main/figures/yolov5.png]
69
-
70
- [Insert YOLOv5-Face results image here from the provided link: https://github.com/IS2AI/TFW/blob/main/figures/yolov5_face.png] -->
 
1
+ ---
2
+ license: mit
3
+ task_categories:
4
+ - object-detection
5
+ tags:
6
+ - thermal faces
7
+ - thermal facial landmarks
8
+ pretty_name: tfw
9
+ size_categories:
10
+ - 1K<n<10K
11
+ ---
12
  # TFW: Annotated Thermal Faces in the Wild Dataset Card
13
+ **Paper:** [TFW: Annotated Thermal Faces in the Wild Dataset (IEEE)](https://ieeexplore.ieee.org/abstract/document/9781417)
14
 
15
+ **GitHub:** [https://github.com/IS2AI/TFW](https://github.com/IS2AI/TFW)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
16
 
17
+ **Description:**
18
+ The TFW dataset provides thermal images collected in a controlled indoor (`c-indoor`), semi-controlled indoor (`s-indoor`), and
19
+ uncontrolled outdoor (`u-outdoor`) environments. The dataset was manually annotated with facial bounding boxes and 5-point facial landmarks.
20
+ | Environment | Subjects | Images | Labeled faces | Visual pair|
21
+ | ---:| :---: | :---: | :---: | :---: |
22
+ | c-indoor | 142 | 5,112 | 5,112 | yes |
23
+ | s-indoor | 9 | 780 | 1,748 | yes |
24
+ | u-outdoor | 15 | 4,090 | 9,649 | no |
25
+ | combined | 147 | 9,982 | 16,509 | yes & no |
26
 
27
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/67056b2e6409e548690b1b6f/uWWvF2xxBfy8jNxaP2Kwk.png)
28
 
29
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
30
  **Citation:**
 
31
  ```bibtex
32
  @ARTICLE{9781417,
33
  author={Kuzdeuov, Askat and Aubakirova, Dana and Koishigarina, Darina and Varol, Huseyin Atakan},
 
38
  number={},
39
  pages={2084-2094},
40
  doi={10.1109/TIFS.2022.3177949}}
41
+ ```