dearsky/wifi-densepose
1
0
Fork 0
Code Issues 20 Pull Requests 5 Actions Packages Projects Releases 1 Wiki Activity

Add WiFi DensePose implementation and results

Browse Source 
- Implemented the WiFi DensePose model in PyTorch, including CSI phase processing, modality translation, and DensePose prediction heads.
- Added a comprehensive training utility for the model, including loss functions and training steps.
- Created a CSV file to document hardware specifications, architecture details, training parameters, performance metrics, and advantages of the model.
This commit is contained in:
rUv
2025-06-07 05:23:07 +00:00
parent 8227a70c31
commit f3c77b1750
66 changed files with 19849 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

61
references/wifi_densepose_results.csv Normal file
View File

@@ -0,0 +1,61 @@
Category,Metric,Value,Unit,Description
Hardware,WiFi_Transmitters,3,count,Number of WiFi transmitter antennas
Hardware,WiFi_Receivers,3,count,Number of WiFi receiver antennas
Hardware,Frequency_Range,2.4GHz ± 20MHz,frequency,Operating frequency range
Hardware,Subcarriers,30,count,Number of subcarrier frequencies
Hardware,Sampling_Rate,100,Hz,CSI data sampling rate
Hardware,Total_Cost,30,USD,Hardware cost using TP-Link AC1750 routers
Architecture,Input_Amplitude_Shape,150x3x3,tensor,CSI amplitude input dimensions
Architecture,Input_Phase_Shape,150x3x3,tensor,CSI phase input dimensions
Architecture,Output_Feature_Shape,3x720x1280,tensor,Spatial feature map dimensions
Architecture,Body_Parts,24,count,Number of body parts detected
Architecture,Keypoints,17,count,Number of keypoints tracked (COCO format)
Training,Learning_Rate,0.001,rate,Initial learning rate
Training,Batch_Size,16,count,Training batch size
Training,Total_Iterations,145000,count,Total training iterations
Training,Lambda_DensePose,0.6,weight,DensePose loss weight
Training,Lambda_Keypoint,0.3,weight,Keypoint loss weight
Training,Lambda_Transfer,0.1,weight,Transfer learning loss weight
Performance,WiFi_Same_Layout_AP,43.5,AP,AP for WiFi_Same_Layout
Performance,WiFi_Same_Layout_AP@50,87.2,AP,AP@50 for WiFi_Same_Layout
Performance,WiFi_Same_Layout_AP@75,44.6,AP,AP@75 for WiFi_Same_Layout
Performance,WiFi_Same_Layout_AP-m,38.1,AP,AP-m for WiFi_Same_Layout
Performance,WiFi_Same_Layout_AP-l,46.4,AP,AP-l for WiFi_Same_Layout
Performance,WiFi_Same_Layout_dpAP_GPS,45.3,AP,dpAP_GPS for WiFi_Same_Layout
Performance,WiFi_Same_Layout_dpAP_GPS@50,79.3,AP,dpAP_GPS@50 for WiFi_Same_Layout
Performance,WiFi_Same_Layout_dpAP_GPS@75,47.7,AP,dpAP_GPS@75 for WiFi_Same_Layout
Performance,WiFi_Same_Layout_dpAP_GPSm,43.2,AP,dpAP_GPSm for WiFi_Same_Layout
Performance,WiFi_Same_Layout_dpAP_GPSm@50,77.4,AP,dpAP_GPSm@50 for WiFi_Same_Layout
Performance,WiFi_Same_Layout_dpAP_GPSm@75,45.5,AP,dpAP_GPSm@75 for WiFi_Same_Layout
Performance,Image_Same_Layout_AP,84.7,AP,AP for Image_Same_Layout
Performance,Image_Same_Layout_AP@50,94.4,AP,AP@50 for Image_Same_Layout
Performance,Image_Same_Layout_AP@75,77.1,AP,AP@75 for Image_Same_Layout
Performance,Image_Same_Layout_AP-m,70.3,AP,AP-m for Image_Same_Layout
Performance,Image_Same_Layout_AP-l,83.8,AP,AP-l for Image_Same_Layout
Performance,Image_Same_Layout_dpAP_GPS,81.8,AP,dpAP_GPS for Image_Same_Layout
Performance,Image_Same_Layout_dpAP_GPS@50,93.7,AP,dpAP_GPS@50 for Image_Same_Layout
Performance,Image_Same_Layout_dpAP_GPS@75,86.2,AP,dpAP_GPS@75 for Image_Same_Layout
Performance,Image_Same_Layout_dpAP_GPSm,84.0,AP,dpAP_GPSm for Image_Same_Layout
Performance,Image_Same_Layout_dpAP_GPSm@50,94.9,AP,dpAP_GPSm@50 for Image_Same_Layout
Performance,Image_Same_Layout_dpAP_GPSm@75,86.8,AP,dpAP_GPSm@75 for Image_Same_Layout
Performance,WiFi_Different_Layout_AP,27.3,AP,AP for WiFi_Different_Layout
Performance,WiFi_Different_Layout_AP@50,51.8,AP,AP@50 for WiFi_Different_Layout
Performance,WiFi_Different_Layout_AP@75,24.2,AP,AP@75 for WiFi_Different_Layout
Performance,WiFi_Different_Layout_AP-m,22.1,AP,AP-m for WiFi_Different_Layout
Performance,WiFi_Different_Layout_AP-l,28.6,AP,AP-l for WiFi_Different_Layout
Performance,WiFi_Different_Layout_dpAP_GPS,25.4,AP,dpAP_GPS for WiFi_Different_Layout
Performance,WiFi_Different_Layout_dpAP_GPS@50,50.2,AP,dpAP_GPS@50 for WiFi_Different_Layout
Performance,WiFi_Different_Layout_dpAP_GPS@75,24.7,AP,dpAP_GPS@75 for WiFi_Different_Layout
Performance,WiFi_Different_Layout_dpAP_GPSm,23.2,AP,dpAP_GPSm for WiFi_Different_Layout
Performance,WiFi_Different_Layout_dpAP_GPSm@50,47.4,AP,dpAP_GPSm@50 for WiFi_Different_Layout
Performance,WiFi_Different_Layout_dpAP_GPSm@75,26.5,AP,dpAP_GPSm@75 for WiFi_Different_Layout
Ablation,Amplitude_Only_AP,39.5,AP,Performance with amplitude only
Ablation,Plus_Phase_AP,40.3,AP,Performance adding phase information
Ablation,Plus_Keypoints_AP,42.9,AP,Performance adding keypoint supervision
Ablation,Final_Model_AP,43.5,AP,Performance with transfer learning
Advantages,Through_Walls,Yes,boolean,Can detect through walls and obstacles
Advantages,Privacy_Preserving,Yes,boolean,No visual recording required
Advantages,Lighting_Independent,Yes,boolean,Works in complete darkness
Advantages,Low_Cost,Yes,boolean,Uses standard WiFi equipment
Advantages,Real_Time,Yes,boolean,Multiple frames per second
Advantages,Multiple_People,Yes,boolean,Can track multiple people simultaneously
1 Category Metric Value Unit Description
2 Hardware WiFi_Transmitters 3 count Number of WiFi transmitter antennas
3 Hardware WiFi_Receivers 3 count Number of WiFi receiver antennas
4 Hardware Frequency_Range 2.4GHz ± 20MHz frequency Operating frequency range
5 Hardware Subcarriers 30 count Number of subcarrier frequencies
6 Hardware Sampling_Rate 100 Hz CSI data sampling rate
7 Hardware Total_Cost 30 USD Hardware cost using TP-Link AC1750 routers
8 Architecture Input_Amplitude_Shape 150x3x3 tensor CSI amplitude input dimensions
9 Architecture Input_Phase_Shape 150x3x3 tensor CSI phase input dimensions
10 Architecture Output_Feature_Shape 3x720x1280 tensor Spatial feature map dimensions
11 Architecture Body_Parts 24 count Number of body parts detected
12 Architecture Keypoints 17 count Number of keypoints tracked (COCO format)
13 Training Learning_Rate 0.001 rate Initial learning rate
14 Training Batch_Size 16 count Training batch size
15 Training Total_Iterations 145000 count Total training iterations
16 Training Lambda_DensePose 0.6 weight DensePose loss weight
17 Training Lambda_Keypoint 0.3 weight Keypoint loss weight
18 Training Lambda_Transfer 0.1 weight Transfer learning loss weight
19 Performance WiFi_Same_Layout_AP 43.5 AP AP for WiFi_Same_Layout
20 Performance WiFi_Same_Layout_AP@50 87.2 AP AP@50 for WiFi_Same_Layout
21 Performance WiFi_Same_Layout_AP@75 44.6 AP AP@75 for WiFi_Same_Layout
22 Performance WiFi_Same_Layout_AP-m 38.1 AP AP-m for WiFi_Same_Layout
23 Performance WiFi_Same_Layout_AP-l 46.4 AP AP-l for WiFi_Same_Layout
24 Performance WiFi_Same_Layout_dpAP_GPS 45.3 AP dpAP_GPS for WiFi_Same_Layout
25 Performance WiFi_Same_Layout_dpAP_GPS@50 79.3 AP dpAP_GPS@50 for WiFi_Same_Layout
26 Performance WiFi_Same_Layout_dpAP_GPS@75 47.7 AP dpAP_GPS@75 for WiFi_Same_Layout
27 Performance WiFi_Same_Layout_dpAP_GPSm 43.2 AP dpAP_GPSm for WiFi_Same_Layout
28 Performance WiFi_Same_Layout_dpAP_GPSm@50 77.4 AP dpAP_GPSm@50 for WiFi_Same_Layout
29 Performance WiFi_Same_Layout_dpAP_GPSm@75 45.5 AP dpAP_GPSm@75 for WiFi_Same_Layout
30 Performance Image_Same_Layout_AP 84.7 AP AP for Image_Same_Layout
31 Performance Image_Same_Layout_AP@50 94.4 AP AP@50 for Image_Same_Layout
32 Performance Image_Same_Layout_AP@75 77.1 AP AP@75 for Image_Same_Layout
33 Performance Image_Same_Layout_AP-m 70.3 AP AP-m for Image_Same_Layout
34 Performance Image_Same_Layout_AP-l 83.8 AP AP-l for Image_Same_Layout
35 Performance Image_Same_Layout_dpAP_GPS 81.8 AP dpAP_GPS for Image_Same_Layout
36 Performance Image_Same_Layout_dpAP_GPS@50 93.7 AP dpAP_GPS@50 for Image_Same_Layout
37 Performance Image_Same_Layout_dpAP_GPS@75 86.2 AP dpAP_GPS@75 for Image_Same_Layout
38 Performance Image_Same_Layout_dpAP_GPSm 84.0 AP dpAP_GPSm for Image_Same_Layout
39 Performance Image_Same_Layout_dpAP_GPSm@50 94.9 AP dpAP_GPSm@50 for Image_Same_Layout
40 Performance Image_Same_Layout_dpAP_GPSm@75 86.8 AP dpAP_GPSm@75 for Image_Same_Layout
41 Performance WiFi_Different_Layout_AP 27.3 AP AP for WiFi_Different_Layout
42 Performance WiFi_Different_Layout_AP@50 51.8 AP AP@50 for WiFi_Different_Layout
43 Performance WiFi_Different_Layout_AP@75 24.2 AP AP@75 for WiFi_Different_Layout
44 Performance WiFi_Different_Layout_AP-m 22.1 AP AP-m for WiFi_Different_Layout
45 Performance WiFi_Different_Layout_AP-l 28.6 AP AP-l for WiFi_Different_Layout
46 Performance WiFi_Different_Layout_dpAP_GPS 25.4 AP dpAP_GPS for WiFi_Different_Layout
47 Performance WiFi_Different_Layout_dpAP_GPS@50 50.2 AP dpAP_GPS@50 for WiFi_Different_Layout
48 Performance WiFi_Different_Layout_dpAP_GPS@75 24.7 AP dpAP_GPS@75 for WiFi_Different_Layout
49 Performance WiFi_Different_Layout_dpAP_GPSm 23.2 AP dpAP_GPSm for WiFi_Different_Layout
50 Performance WiFi_Different_Layout_dpAP_GPSm@50 47.4 AP dpAP_GPSm@50 for WiFi_Different_Layout
51 Performance WiFi_Different_Layout_dpAP_GPSm@75 26.5 AP dpAP_GPSm@75 for WiFi_Different_Layout
52 Ablation Amplitude_Only_AP 39.5 AP Performance with amplitude only
53 Ablation Plus_Phase_AP 40.3 AP Performance adding phase information
54 Ablation Plus_Keypoints_AP 42.9 AP Performance adding keypoint supervision
55 Ablation Final_Model_AP 43.5 AP Performance with transfer learning
56 Advantages Through_Walls Yes boolean Can detect through walls and obstacles
57 Advantages Privacy_Preserving Yes boolean No visual recording required
58 Advantages Lighting_Independent Yes boolean Works in complete darkness
59 Advantages Low_Cost Yes boolean Uses standard WiFi equipment
60 Advantages Real_Time Yes boolean Multiple frames per second
61 Advantages Multiple_People Yes boolean Can track multiple people simultaneously