subhash25rawat/custom-hair 🖼️ → 🖼️
Performance
45.2sTypical run time
~336sCold start (first call)
1.1KTotal runs
About
Customise your hair with AI. Swap hair with anyone, copy anyone's hair color.
Example Output
Output
Performance Metrics
45.17s
Prediction Time
335.56s
Total Time
All Input Parameters
{
"face_image": "https://replicate.delivery/pbxt/MPV6rRRX2WVexpahBMcYL9qRnFQg8XsISFV1CUgwvcrUQ4WJ/29995.png",
"hair_image": "https://replicate.delivery/pbxt/MPV6s0mFWBBE6SjmhRADZSv02YWpKF4KYk5KisZDutqTILI3/06041.png",
"color_image": "https://replicate.delivery/pbxt/MPV6rd7oS2paLzlwjtFJ1tK8ArjLUK6RCu7QVUO3hLkR6AOH/29995.png"
}
Input Parameters
- face_image (required)
- Image of the person whose hair style is to be changed
- hair_image (required)
- Image of the person whose hair style is to be copied
- color_image
- Image of the person whose hair color is to be copied
Output Schema
Output
Example Execution Logs
Original image size: 1024x1024 Resized image size: 720x720 Image pasted at: 0, 0 Original image size: 1024x1024 Resized image size: 720x720 Image pasted at: 0, 0 Original image size: 1024x1024 Resized image size: 720x720 Image pasted at: 0, 0 Number of faces detected: 1 Number of faces detected: 1 Loading StyleGAN2 from checkpoint: pretrained_models/StyleGAN/ffhq.pt Downloading: "https://download.pytorch.org/models/resnet18-5c106cde.pth" to /root/.cache/torch/hub/checkpoints/resnet18-5c106cde.pth 0%| | 0.00/44.7M [00:00<?, ?B/s] 21%|██ | 9.45M/44.7M [00:00<00:00, 99.1MB/s] 78%|███████▊ | 34.7M/44.7M [00:00<00:00, 196MB/s] 100%|██████████| 44.7M/44.7M [00:00<00:00, 195MB/s] Loading e4e over the pSp framework from checkpoint: pretrained_models/encoder4editing/e4e_ffhq_encode.pt Network [SPADEGenerator] was created. Total number of parameters: 266.9 million. To see the architecture, do print(network). 0%| | 0.00/335M [00:00<?, ?iB/s] 2%|▉ | 8.03M/335M [00:00<00:04, 83.5MiB/s] 5%|█▉ | 17.4M/335M [00:00<00:03, 91.8MiB/s] 9%|███▍ | 29.3M/335M [00:00<00:02, 107MiB/s] 13%|█████▏ | 44.4M/335M [00:00<00:02, 127MiB/s] 17%|██████▌ | 56.5M/335M [00:00<00:02, 105MiB/s] 20%|███████▌ | 67.0M/335M [00:00<00:03, 85.9MiB/s] 24%|████████▉ | 79.0M/335M [00:00<00:02, 96.2MiB/s] 28%|███████████ | 94.4M/335M [00:00<00:02, 114MiB/s] 32%|████████████▋ | 106M/335M [00:01<00:02, 110MiB/s] 35%|██████████████ | 117M/335M [00:01<00:02, 105MiB/s] 38%|███████████████▏ | 128M/335M [00:01<00:02, 105MiB/s] 43%|█████████████████ | 143M/335M [00:01<00:01, 121MiB/s] 46%|██████████████████▌ | 155M/335M [00:01<00:01, 121MiB/s] 50%|███████████████████▉ | 167M/335M [00:01<00:01, 115MiB/s] 53%|█████████████████████▎ | 178M/335M [00:01<00:01, 101MiB/s] 56%|█████████████████████▉ | 188M/335M [00:01<00:01, 96.5MiB/s] 59%|███████████████████████ | 198M/335M [00:01<00:01, 96.4MiB/s] 62%|████████████████████████▏ | 207M/335M [00:02<00:01, 94.3MiB/s] 65%|█████████████████████████▏ | 217M/335M [00:02<00:01, 95.3MiB/s] 68%|███████████████████████████▏ | 227M/335M [00:02<00:01, 101MiB/s] 71%|███████████████████████████▋ | 237M/335M [00:02<00:01, 97.8MiB/s] 74%|█████████████████████████████▌ | 247M/335M [00:02<00:00, 101MiB/s] 77%|██████████████████████████████▊ | 258M/335M [00:02<00:00, 104MiB/s] 80%|███████████████████████████████▏ | 268M/335M [00:02<00:00, 89.6MiB/s] 83%|████████████████████████████████▎ | 277M/335M [00:02<00:00, 72.5MiB/s] 86%|█████████████████████████████████▍ | 287M/335M [00:03<00:00, 80.4MiB/s] 89%|██████████████████████████████████▉ | 299M/335M [00:03<00:00, 92.7MiB/s] 93%|████████████████████████████████████▏ | 310M/335M [00:03<00:00, 98.3MiB/s] 96%|██████████████████████████████████████▌ | 322M/335M [00:03<00:00, 106MiB/s] 100%|████████████████████████████████████████| 335M/335M [00:03<00:00, 102MiB/s] 0%| | 0/1 [00:00<?, ?it/s] 0%| | 0/1 [00:04<?, ?it/s]
Version Details
- Version ID
79fcdf78d664c6f7bfd2468fdd3db2ba514bf5d777300f951f5e4b19cdbdbb5f- Version Created
- January 29, 2025