Do not touch the image if the size is good

app.py

@@ -14,14 +14,11 @@ import uuid
 import random
 from huggingface_hub import hf_hub_download
 import spaces
-#gradio.helpers.CACHED_FOLDER = '/data/cache'
 
 pipe = StableVideoDiffusionPipeline.from_pretrained(
     "vdo/stable-video-diffusion-img2vid-xt-1-1", torch_dtype=torch.float16, variant="fp16"
 )
 pipe.to("cuda")
-#pipe.unet = torch.compile(pipe.unet, mode="reduce-overhead", fullgraph=True)
-#pipe.vae = torch.compile(pipe.vae, mode="reduce-overhead", fullgraph=True)
 
 max_64_bit_int = 2**63 - 1
 
@@ -51,7 +48,6 @@ def sample(
 
     frames = pipe(image, decode_chunk_size=decoding_t, generator=generator, motion_bucket_id=motion_bucket_id, noise_aug_strength=0.1, num_frames=25).frames[0]
     export_to_video(frames, video_path, fps=fps_id)
-    torch.manual_seed(seed)
     
     return video_path, frames, seed
 
@@ -60,6 +56,10 @@ def resize_image(image, output_size=(1024, 576)):
     target_aspect = output_size[0] / output_size[1]  # Aspect ratio of the desired size
     image_aspect = image.width / image.height  # Aspect ratio of the original image
 
+    # Do not touch the image if the size is good
+    if image.width == output_size[0] and image.height == output_size[1]:
+        return image
+
     # Resize then crop if the original image is larger
     if image_aspect > target_aspect:
         # Resize the image to match the target height, maintaining aspect ratio
@@ -107,5 +107,4 @@ with gr.Blocks() as demo:
   generate_btn.click(fn=sample, inputs=[image, seed, randomize_seed, motion_bucket_id, fps_id], outputs=[video, gallery, seed], api_name="video")
 
 if __name__ == "__main__":
-    #demo.queue(max_size=20, api_open=False)
     demo.launch(share=True, show_api=False)