app.py

import gradio as gr
import torch
from PIL import Image
from transformers import AutoTokenizer, AutoModelForCausalLM
import cv2
import numpy as np
from typing import Optional
import tempfile
import os
import spaces

MID = "apple/FastVLM-7B"
IMAGE_TOKEN_INDEX = -200

# Initialize model variables
tok = None
model = None

def load_model():
    global tok, model
    if tok is None or model is None:
        print("Loading FastVLM model...")
        tok = AutoTokenizer.from_pretrained(MID, trust_remote_code=True)
        model = AutoModelForCausalLM.from_pretrained(
            MID,
            torch_dtype=torch.float16,
            device_map="cuda",
            trust_remote_code=True,
        )
        print("Model loaded successfully!")
    return tok, model

def extract_frames(video_path: str, num_frames: int = 8, sampling_method: str = "uniform"):
    """Extract frames from video"""
    cap = cv2.VideoCapture(video_path)
    total_frames = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
    
    if total_frames == 0:
        cap.release()
        return []
    
    frames = []
    
    if sampling_method == "uniform":
        # Uniform sampling
        indices = np.linspace(0, total_frames - 1, num_frames, dtype=int)
    elif sampling_method == "first":
        # Take first N frames
        indices = list(range(min(num_frames, total_frames)))
    elif sampling_method == "last":
        # Take last N frames
        start = max(0, total_frames - num_frames)
        indices = list(range(start, total_frames))
    else:  # middle
        # Take frames from the middle
        start = max(0, (total_frames - num_frames) // 2)
        indices = list(range(start, min(start + num_frames, total_frames)))
    
    for idx in indices:
        cap.set(cv2.CAP_PROP_POS_FRAMES, idx)
        ret, frame = cap.read()
        if ret:
            # Convert BGR to RGB
            frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
            frames.append(Image.fromarray(frame_rgb))
    
    cap.release()
    return frames

@spaces.GPU(duration=60)
def caption_frame(image: Image.Image, prompt: str) -> str:
    """Generate caption for a single frame"""
    # Load model on GPU
    tok, model = load_model()
    # Build chat with custom prompt
    messages = [
        {"role": "user", "content": f"<image>\n{prompt}"}
    ]
    rendered = tok.apply_chat_template(
        messages, add_generation_prompt=True, tokenize=False
    )
    pre, post = rendered.split("<image>", 1)
    
    # Tokenize the text around the image token
    pre_ids = tok(pre, return_tensors="pt", add_special_tokens=False).input_ids
    post_ids = tok(post, return_tensors="pt", add_special_tokens=False).input_ids
    
    # Splice in the IMAGE token id
    img_tok = torch.tensor([[IMAGE_TOKEN_INDEX]], dtype=pre_ids.dtype)
    input_ids = torch.cat([pre_ids, img_tok, post_ids], dim=1).to(model.device)
    attention_mask = torch.ones_like(input_ids, device=model.device)
    
    # Preprocess image
    px = model.get_vision_tower().image_processor(images=image, return_tensors="pt")["pixel_values"]
    px = px.to(model.device, dtype=model.dtype)
    
    # Generate
    with torch.no_grad():
        out = model.generate(
            inputs=input_ids,
            attention_mask=attention_mask,
            images=px,
            max_new_tokens=15,
            temperature=0.7,
            do_sample=True,
        )
    
    caption = tok.decode(out[0], skip_special_tokens=True)
    # Extract only the generated part
    if prompt in caption:
        caption = caption.split(prompt)[-1].strip()
    
    return caption

def process_video(
    video_path: str,
    num_frames: int,
    sampling_method: str,
    caption_mode: str,
    custom_prompt: str,
    progress=gr.Progress()
) -> tuple:
    """Process video and generate captions"""
    
    if not video_path:
        return "Please upload a video first.", None
    
    progress(0, desc="Extracting frames...")
    frames = extract_frames(video_path, num_frames, sampling_method)
    
    if not frames:
        return "Failed to extract frames from video.", None
    
    # Use brief one-sentence prompt for faster processing
    prompt = "Provide a brief one-sentence description of what's happening in this image."
    
    captions = []
    frame_previews = []
    
    for i, frame in enumerate(frames):
        progress((i + 1) / (len(frames) + 1), desc=f"Analyzing frame {i + 1}/{len(frames)}...")
        caption = caption_frame(frame, prompt)
        captions.append(f"Frame {i + 1}: {caption}")
        frame_previews.append(frame)
    
    progress(1.0, desc="Generating summary...")
    
    # Combine captions into a simple narrative
    full_caption = "\n".join(captions)
    
    # Generate overall summary if multiple frames
    if len(frames) > 1:
        video_summary = f"Analyzed {len(frames)} frames:\n\n{full_caption}"
    else:
        video_summary = f"Video Analysis:\n\n{full_caption}"
    
    return video_summary, frame_previews

# Create the Gradio interface
# Create custom Apple-inspired theme
class AppleTheme(gr.themes.Base):
    def __init__(self):
        super().__init__(
            primary_hue=gr.themes.colors.blue,
            secondary_hue=gr.themes.colors.gray,
            neutral_hue=gr.themes.colors.gray,
            spacing_size=gr.themes.sizes.spacing_md,
            radius_size=gr.themes.sizes.radius_md,
            text_size=gr.themes.sizes.text_md,
            font=[
                gr.themes.GoogleFont("Inter"),
                "-apple-system",
                "BlinkMacSystemFont",
                "SF Pro Display",
                "SF Pro Text",
                "Helvetica Neue",
                "Helvetica",
                "Arial",
                "sans-serif"
            ],
            font_mono=[
                gr.themes.GoogleFont("SF Mono"),
                "ui-monospace",
                "Consolas",
                "monospace"
            ]
        )
        super().set(
            # Core colors
            body_background_fill="*neutral_50",
            body_background_fill_dark="*neutral_950",
            button_primary_background_fill="*primary_500",
            button_primary_background_fill_hover="*primary_600",
            button_primary_text_color="white",
            button_primary_border_color="*primary_500",
            
            # Shadows
            button_shadow="0 2px 8px rgba(0, 0, 0, 0.04)",
            button_shadow_hover="0 4px 12px rgba(0, 0, 0, 0.08)",
            block_shadow="0 4px 12px rgba(0, 0, 0, 0.08)",
            
            # Border radius
            button_large_radius="8px",
            button_small_radius="6px",
            block_radius="12px",
            container_radius="12px",
            
            # Borders
            block_border_width="1px",
            block_border_color="*neutral_200",
            input_border_width="1px",
            input_border_color="*neutral_300",
            input_border_color_focus="*primary_500",
            
            # Text
            block_title_text_weight="600",
            block_label_text_weight="500",
            block_label_text_size="13px",
            block_label_text_color="*neutral_600",
            body_text_color="*neutral_900",
            
            # Spacing
            layout_gap="16px",
            block_padding="20px",
            
            # Specific components
            chatbot_code_background_color="*neutral_100",
            slider_color="*primary_500",
            
            # Transitions
            button_transition="all 0.2s cubic-bezier(0.4, 0, 0.2, 1)",
        )

# Create the Gradio interface with the custom theme
with gr.Blocks(theme=AppleTheme()) as demo:
    gr.Markdown("# 🎬 FastVLM Video Captioning")
    
    with gr.Row():
        # Main video display
        with gr.Column(scale=7):
            video_display = gr.Video(
                label="Video Input",
                autoplay=True,
                loop=True
            )
        
        # Sidebar with chat interface
        with gr.Sidebar(width=400):
            gr.Markdown("## 💬 Video Analysis Chat")
            
            chatbot = gr.Chatbot(
                value=[["Assistant", "Upload a video and I'll analyze it for you!"]],
                height=400,
                elem_classes=["chatbot"]
            )
            
            process_btn = gr.Button("🎯 Analyze Video", variant="primary", size="lg")
            
            with gr.Accordion("🖼️ Analyzed Frames", open=False):
                frame_gallery = gr.Gallery(
                    label="Extracted Frames",
                    show_label=False,
                    columns=2,
                    rows=4,
                    object_fit="contain",
                    height="auto"
                )
    
    # Hidden parameters with default values
    num_frames = gr.State(value=8)
    sampling_method = gr.State(value="uniform")
    caption_mode = gr.State(value="Brief Summary")
    custom_prompt = gr.State(value="")
    
    # Upload handler
    def handle_upload(video, chat_history):
        if video:
            chat_history.append(["User", "Video uploaded"])
            chat_history.append(["Assistant", "Video loaded! Click 'Analyze Video' to generate captions."])
            return video, chat_history
        return None, chat_history
    
    video_display.upload(
        handle_upload,
        inputs=[video_display, chatbot],
        outputs=[video_display, chatbot]
    )
    
    # Modified process function to update chatbot with streaming
    def process_video_with_chat(video_path, num_frames, sampling_method, caption_mode, custom_prompt, chat_history, progress=gr.Progress()):
        if not video_path:
            chat_history.append(["Assistant", "Please upload a video first."])
            yield chat_history, None
            return
        
        chat_history.append(["User", "Analyzing video..."])
        yield chat_history, None
        
        # Extract frames
        progress(0, desc="Extracting frames...")
        frames = extract_frames(video_path, num_frames, sampling_method)
        
        if not frames:
            chat_history.append(["Assistant", "Failed to extract frames from video."])
            yield chat_history, None
            return
        
        # Start streaming response
        chat_history.append(["Assistant", ""])
        prompt = "Provide a brief one-sentence description of what's happening in this image."
        
        captions = []
        for i, frame in enumerate(frames):
            progress((i + 1) / (len(frames) + 1), desc=f"Analyzing frame {i + 1}/{len(frames)}...")
            caption = caption_frame(frame, prompt)
            frame_caption = f"Frame {i + 1}: {caption}\n"
            captions.append(frame_caption)
            
            # Update the last message with accumulated captions
            current_text = "".join(captions)
            chat_history[-1] = ["Assistant", f"Analyzing {len(frames)} frames:\n\n{current_text}"]
            yield chat_history, frames[:i+1]  # Also update frame gallery progressively
        
        progress(1.0, desc="Analysis complete!")
        
        # Final update with complete message
        full_caption = "".join(captions)
        final_message = f"Analyzed {len(frames)} frames:\n\n{full_caption}"
        chat_history[-1] = ["Assistant", final_message]
        yield chat_history, frames
    
    # Process button with streaming
    process_btn.click(
        process_video_with_chat,
        inputs=[video_display, num_frames, sampling_method, caption_mode, custom_prompt, chatbot],
        outputs=[chatbot, frame_gallery],
        show_progress=True
    )
    
    demo.launch()