feat(api): add strength param to inpaint, remove same from upscale

api/onnx_web/pipeline.py

@@ -30,10 +30,11 @@ last_pipeline_instance = None
 last_pipeline_options = (None, None, None)
 last_pipeline_scheduler = None
 
-# from https://www.travelneil.com/stable-diffusion-updates.html
-
 
 def get_latents_from_seed(seed: int, size: Size) -> np.ndarray:
+    '''
+    From https://www.travelneil.com/stable-diffusion-updates.html
+    '''
     # 1 is batch size
     latents_shape = (1, 4, size.height // 8, size.width // 8)
     # Gotta use numpy instead of torch, because torch's randn() doesn't support DML
@@ -147,7 +148,8 @@ def run_inpaint_pipeline(
     mask_image: Image,
     expand: Border,
     noise_source: Any,
-    mask_filter: Any
+    mask_filter: Any,
+    strength: float,
 ):
     pipe = load_pipeline(OnnxStableDiffusionInpaintPipeline,
                          params.model, params.provider, params.scheduler)
@@ -180,6 +182,7 @@ def run_inpaint_pipeline(
         num_inference_steps=params.steps,
         width=size.width,
     ).images[0]
+    image = ImageChops.blend(source_image, image, strength)
 
     if upscale.faces or upscale.scale > 1:
         image = upscale_resrgan(ctx, upscale, image)
@@ -189,17 +192,16 @@ def run_inpaint_pipeline(
 
     print('saved inpaint output: %s' % (dest))
 
+
 def run_upscale_pipeline(
     ctx: ServerContext,
     _params: BaseParams,
     _size: Size,
     output: str,
     upscale: UpscaleParams,
-    source_image: Image,
-    strength: float,
+    source_image: Image
 ):
     image = upscale_resrgan(ctx, upscale, source_image)
-    image = ImageChops.blend(source_image, image, strength)
 
     dest = safer_join(ctx.output_path, output)
     image.save(dest)

api/onnx_web/serve.py

@@ -343,7 +343,7 @@ def img2img():
         'img2img',
         params,
         size,
-        extras=(strength))
+        extras=(strength,))
     print("img2img output: %s" % (output))
 
     source_image.thumbnail((size.width, size.height))
@@ -393,6 +393,11 @@ def inpaint():
     mask_filter = get_from_map(request.args, 'filter', mask_filters, 'none')
     noise_source = get_from_map(
         request.args, 'noise', noise_sources, 'histogram')
+    strength = get_and_clamp_float(
+        request.args,
+        'strength',
+        config_params.get('strength').get('default'),
+        config_params.get('strength').get('max'))
 
     output = make_output_name(
         'inpaint',
@@ -405,6 +410,7 @@ def inpaint():
             expand.bottom,
             mask_filter.__name__,
             noise_source.__name__,
+            strength,
         )
     )
     print("inpaint output: %s" % output)
@@ -423,7 +429,8 @@ def inpaint():
         mask_image,
         expand,
         noise_source,
-        mask_filter)
+        mask_filter,
+        strength)
 
     return jsonify({
         'output': output,
@@ -440,22 +447,15 @@ def upscale():
     params, size = pipeline_from_request()
     upscale = upscale_from_request()
 
-    strength = get_and_clamp_float(
-        request.args,
-        'strength',
-        config_params.get('strength').get('default'),
-        config_params.get('strength').get('max'))
-
     output = make_output_name(
-        'img2img',
+        'upscale',
         params,
-        size,
-        extras=(strength))
-    print("img2img output: %s" % (output))
+        size)
+    print("upscale output: %s" % (output))
 
     source_image.thumbnail((size.width, size.height))
     executor.submit_stored(output, run_upscale_pipeline,
-                           context, params, output, upscale, source_image, strength)
+                           context, params, output, upscale, source_image)
 
     return jsonify({
         'output': output,