Choose the frame format (#11)

- Choose the frame format (ffd4ed8bb7f89e3ce340694c0dc6cb9206079efc)


Co-authored-by: Fabrice TIERCELIN <[email protected]>

app.py

@@ -6,10 +6,9 @@ from pathlib import Path
 from typing import Optional
 
 from diffusers import StableVideoDiffusionPipeline
-from diffusers.utils import load_image, export_to_video
+from diffusers.utils import export_to_video
 from PIL import Image
 
-import uuid
 import random
 import spaces
 
@@ -29,6 +28,7 @@ def sample(
     fps_id: int = 6,
     noise_aug_strength: float = 0.1,
     decoding_t: int = 3,
+    frame_format: str = "webp",
     version: str = "svd_xt",
     device: str = "cuda",
     output_folder: str = "outputs",
@@ -36,7 +36,7 @@ def sample(
     if image.mode == "RGBA":
         image = image.convert("RGB")
         
-    if(randomize_seed):
+    if randomize_seed:
         seed = random.randint(0, max_64_bit_int)
     generator = torch.manual_seed(seed)
     
@@ -47,7 +47,7 @@ def sample(
     frames = pipe(image, decode_chunk_size=decoding_t, generator=generator, motion_bucket_id=motion_bucket_id, noise_aug_strength=noise_aug_strength, num_frames=25).frames[0]
     export_to_video(frames, video_path, fps=fps_id)
     
-    return video_path, frames, seed
+    return video_path, gr.update(label="Generated frames in *." + frame_format + " format", format = frame_format, value = frames), seed
 
 def resize_image(image, output_size=(1024, 576)):
     # Calculate aspect ratios
@@ -96,6 +96,7 @@ with gr.Blocks() as demo:
               motion_bucket_id = gr.Slider(label="Motion bucket id", info="Controls how much motion to add/remove from the image", value=127, minimum=1, maximum=255)
               noise_aug_strength = gr.Slider(label="Noise strength", info="The noise to add", value=0.1, minimum=0, maximum=1, step=0.1)
               decoding_t = gr.Slider(label="Decoding", info="Number of frames decoded at a time; this eats more VRAM; reduce if necessary", value=3, minimum=1, maximum=5, step=1)
+              frame_format = gr.Radio([["*.png", "png"], ["*.webp", "webp"], ["*.jpeg", "jpeg"], ["*.gif", "gif"], ["*.bmp", "bmp"]], label="Image format for result", info="File extention", value="webp", interactive=True)
               seed = gr.Slider(label="Seed", value=42, randomize=True, minimum=0, maximum=max_64_bit_int, step=1)
               randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
 
@@ -106,7 +107,7 @@ with gr.Blocks() as demo:
           gallery = gr.Gallery(label="Generated frames")
       
   image.upload(fn=resize_image, inputs=image, outputs=image, queue=False)
-  generate_btn.click(fn=sample, inputs=[image, seed, randomize_seed, motion_bucket_id, fps_id, noise_aug_strength, decoding_t], outputs=[video, gallery, seed], api_name="video")
+  generate_btn.click(fn=sample, inputs=[image, seed, randomize_seed, motion_bucket_id, fps_id, noise_aug_strength, decoding_t, frame_format], outputs=[video, gallery, seed], api_name="video")
 
 if __name__ == "__main__":
     demo.launch(share=True, show_api=False)