Update app.py

app.py

@@ -1,26 +1,57 @@
-import torch
+import os
+import streamlit as st
 from diffusers import CogVideoXImageToVideoPipeline
 from diffusers.utils import export_to_video, load_image
+import torch
+
+# Streamlit interface for uploading an image and inputting a prompt
+st.title("Image to Video with Hugging Face")
+st.write("Upload an image and provide a prompt to generate a video.")
+
+# File uploader for the input image
+uploaded_file = st.file_uploader("Upload an image (JPG or PNG):", type=["jpg", "jpeg", "png"])
+prompt = st.text_input("Enter your prompt:", "A little girl is riding a bicycle at high speed. Focused, detailed, realistic.")
+
+if uploaded_file and prompt:
+    try:
+        # Save the uploaded file to a temporary location
+        with open("uploaded_image.jpg", "wb") as f:
+            f.write(uploaded_file.read())
+
+        # Load the image
+        image = load_image("uploaded_image.jpg")
+
+        # Initialize the CogVideoX pipeline
+        st.write("Initializing the pipeline...")
+        pipe = CogVideoXImageToVideoPipeline.from_pretrained(
+            "THUDM/CogVideoX1.5-5B-I2V",
+            torch_dtype=torch.bfloat16
+        )
+
+        pipe.enable_sequential_cpu_offload()
+        pipe.vae.enable_tiling()
+        pipe.vae.enable_slicing()
+
+        # Generate the video
+        st.write("Generating video... this may take a while.")
+        video_frames = pipe(
+            prompt=prompt,
+            image=image,
+            num_videos_per_prompt=1,
+            num_inference_steps=50,
+            num_frames=81,
+            guidance_scale=6,
+            generator=torch.Generator(device="cuda").manual_seed(42),
+        ).frames[0]
+
+        # Export the video
+        video_path = "output.mp4"
+        export_to_video(video_frames, video_path, fps=8)
+
+        # Display the video in Streamlit
+        st.video(video_path)
 
-prompt = "A little girl is riding a bicycle at high speed. Focused, detailed, realistic."
-image = load_image(image="input.jpg")
-pipe = CogVideoXImageToVideoPipeline.from_pretrained(
-    "THUDM/CogVideoX1.5-5B-I2V",
-    torch_dtype=torch.bfloat16
-)
-
-pipe.enable_sequential_cpu_offload()
-pipe.vae.enable_tiling()
-pipe.vae.enable_slicing()
-
-video = pipe(
-    prompt=prompt,
-    image=image,
-    num_videos_per_prompt=1,
-    num_inference_steps=50,
-    num_frames=81,
-    guidance_scale=6,
-    generator=torch.Generator(device="cuda").manual_seed(42),
-).frames[0]
-
-export_to_video(video, "output.mp4", fps=8)
+    except Exception as e:
+        st.error(f"An error occurred: {e}")
+else:
+    st.write("Please upload an image and provide a prompt to get started.")