Spaces:
Runtime error
Runtime error
RamAnanth1
commited on
Commit
•
efe3c52
1
Parent(s):
ad64d06
Update app.py
Browse files
app.py
CHANGED
@@ -14,19 +14,30 @@ from annotator.openpose import apply_openpose
|
|
14 |
from cldm.model import create_model, load_state_dict
|
15 |
|
16 |
from huggingface_hub import hf_hub_url, cached_download
|
|
|
17 |
REPO_ID = "lllyasviel/ControlNet"
|
18 |
-
|
|
|
19 |
|
20 |
-
|
21 |
-
|
22 |
-
hf_hub_url(REPO_ID,
|
23 |
), location='cpu'))
|
24 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
25 |
|
26 |
def process(input_image, prompt, input_control, a_prompt, n_prompt, num_samples, image_resolution, ddim_steps, scale, seed, eta, low_threshold, high_threshold):
|
27 |
# TODO: Add other control tasks
|
28 |
-
|
29 |
-
|
|
|
|
|
30 |
|
31 |
def process_canny(input_image, prompt, a_prompt, n_prompt, num_samples, image_resolution, ddim_steps, scale, seed, eta, low_threshold, high_threshold):
|
32 |
with torch.no_grad():
|
@@ -42,24 +53,24 @@ def process_canny(input_image, prompt, a_prompt, n_prompt, num_samples, image_re
|
|
42 |
|
43 |
seed_everything(seed)
|
44 |
|
45 |
-
cond = {"c_concat": [control], "c_crossattn": [
|
46 |
-
un_cond = {"c_concat": [control], "c_crossattn": [
|
47 |
shape = (4, H // 8, W // 8)
|
48 |
|
49 |
-
samples, intermediates =
|
50 |
shape, cond, verbose=False, eta=eta,
|
51 |
unconditional_guidance_scale=scale,
|
52 |
unconditional_conditioning=un_cond)
|
53 |
-
x_samples =
|
54 |
x_samples = (einops.rearrange(x_samples, 'b c h w -> b h w c') * 127.5 + 127.5).cpu().numpy().clip(0, 255).astype(np.uint8)
|
55 |
|
56 |
results = [x_samples[i] for i in range(num_samples)]
|
57 |
return [255 - detected_map] + results
|
58 |
|
59 |
-
def process_pose(input_image, prompt, a_prompt, n_prompt, num_samples, image_resolution,
|
60 |
with torch.no_grad():
|
61 |
input_image = HWC3(input_image)
|
62 |
-
detected_map, _ = apply_openpose(resize_image(input_image,
|
63 |
detected_map = HWC3(detected_map)
|
64 |
img = resize_image(input_image, image_resolution)
|
65 |
H, W, C = img.shape
|
@@ -72,15 +83,15 @@ def process_pose(input_image, prompt, a_prompt, n_prompt, num_samples, image_res
|
|
72 |
|
73 |
seed_everything(seed)
|
74 |
|
75 |
-
cond = {"c_concat": [control], "c_crossattn": [
|
76 |
-
un_cond = {"c_concat": [control], "c_crossattn": [
|
77 |
shape = (4, H // 8, W // 8)
|
78 |
|
79 |
-
samples, intermediates =
|
80 |
shape, cond, verbose=False, eta=eta,
|
81 |
unconditional_guidance_scale=scale,
|
82 |
unconditional_conditioning=un_cond)
|
83 |
-
x_samples =
|
84 |
x_samples = (einops.rearrange(x_samples, 'b c h w -> b h w c') * 127.5 + 127.5).cpu().numpy().clip(0, 255).astype(np.uint8)
|
85 |
|
86 |
results = [x_samples[i] for i in range(num_samples)]
|
|
|
14 |
from cldm.model import create_model, load_state_dict
|
15 |
|
16 |
from huggingface_hub import hf_hub_url, cached_download
|
17 |
+
|
18 |
REPO_ID = "lllyasviel/ControlNet"
|
19 |
+
canny_checkpoint = "models/control_sd15_canny.pth"
|
20 |
+
pose_checkpoint = "models/control_sd15_openpose.pth"
|
21 |
|
22 |
+
canny_model = create_model('./models/cldm_v15.yaml')
|
23 |
+
canny_model.load_state_dict(load_state_dict(cached_download(
|
24 |
+
hf_hub_url(REPO_ID, canny_checkpoint)
|
25 |
), location='cpu'))
|
26 |
+
ddim_sampler_canny = DDIMSampler(canny_model)
|
27 |
+
|
28 |
+
|
29 |
+
pose_model = create_model('./models/cldm_v15.yaml')
|
30 |
+
pose_model.load_state_dict(load_state_dict(cached_download(
|
31 |
+
hf_hub_url(REPO_ID, pose_checkpoint)
|
32 |
+
), location='cpu'))
|
33 |
+
ddim_sampler_pose = DDIMSampler(pose_model)
|
34 |
|
35 |
def process(input_image, prompt, input_control, a_prompt, n_prompt, num_samples, image_resolution, ddim_steps, scale, seed, eta, low_threshold, high_threshold):
|
36 |
# TODO: Add other control tasks
|
37 |
+
if input_control == "Canny Edge Map":
|
38 |
+
return process_canny(input_image, prompt, a_prompt, n_prompt, num_samples, image_resolution, ddim_steps, scale, seed, eta, low_threshold, high_threshold)
|
39 |
+
else:
|
40 |
+
return process_pose(input_image, prompt, a_prompt, n_prompt, num_samples, image_resolution, ddim_steps, scale, seed, eta)
|
41 |
|
42 |
def process_canny(input_image, prompt, a_prompt, n_prompt, num_samples, image_resolution, ddim_steps, scale, seed, eta, low_threshold, high_threshold):
|
43 |
with torch.no_grad():
|
|
|
53 |
|
54 |
seed_everything(seed)
|
55 |
|
56 |
+
cond = {"c_concat": [control], "c_crossattn": [canny_model.get_learned_conditioning([prompt + ', ' + a_prompt] * num_samples)]}
|
57 |
+
un_cond = {"c_concat": [control], "c_crossattn": [canny_model.get_learned_conditioning([n_prompt] * num_samples)]}
|
58 |
shape = (4, H // 8, W // 8)
|
59 |
|
60 |
+
samples, intermediates = ddim_sampler_canny.sample(ddim_steps, num_samples,
|
61 |
shape, cond, verbose=False, eta=eta,
|
62 |
unconditional_guidance_scale=scale,
|
63 |
unconditional_conditioning=un_cond)
|
64 |
+
x_samples = canny_model.decode_first_stage(samples)
|
65 |
x_samples = (einops.rearrange(x_samples, 'b c h w -> b h w c') * 127.5 + 127.5).cpu().numpy().clip(0, 255).astype(np.uint8)
|
66 |
|
67 |
results = [x_samples[i] for i in range(num_samples)]
|
68 |
return [255 - detected_map] + results
|
69 |
|
70 |
+
def process_pose(input_image, prompt, a_prompt, n_prompt, num_samples, image_resolution, ddim_steps, scale, seed, eta):
|
71 |
with torch.no_grad():
|
72 |
input_image = HWC3(input_image)
|
73 |
+
detected_map, _ = apply_openpose(resize_image(input_image, image_resolution))
|
74 |
detected_map = HWC3(detected_map)
|
75 |
img = resize_image(input_image, image_resolution)
|
76 |
H, W, C = img.shape
|
|
|
83 |
|
84 |
seed_everything(seed)
|
85 |
|
86 |
+
cond = {"c_concat": [control], "c_crossattn": [pose_model.get_learned_conditioning([prompt + ', ' + a_prompt] * num_samples)]}
|
87 |
+
un_cond = {"c_concat": [control], "c_crossattn": [pose_model.get_learned_conditioning([n_prompt] * num_samples)]}
|
88 |
shape = (4, H // 8, W // 8)
|
89 |
|
90 |
+
samples, intermediates = ddim_sampler_pose.sample(ddim_steps, num_samples,
|
91 |
shape, cond, verbose=False, eta=eta,
|
92 |
unconditional_guidance_scale=scale,
|
93 |
unconditional_conditioning=un_cond)
|
94 |
+
x_samples = pose_model.decode_first_stage(samples)
|
95 |
x_samples = (einops.rearrange(x_samples, 'b c h w -> b h w c') * 127.5 + 127.5).cpu().numpy().clip(0, 255).astype(np.uint8)
|
96 |
|
97 |
results = [x_samples[i] for i in range(num_samples)]
|