File size: 8,974 Bytes
06a6e91
eacc26a
6123d43
 
 
06a6e91
b1fb753
 
 
 
 
 
b8aa7c5
 
4eada0f
b8aa7c5
 
 
b1fb753
6a9e2b5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
a361f34
06a6e91
eacc26a
06a6e91
eacc26a
 
 
 
836daad
2cde650
eacc26a
 
 
b8aa7c5
88d63cf
 
 
 
 
eacc26a
 
 
88d63cf
836daad
eacc26a
2cde650
eacc26a
 
a6b51c5
eacc26a
 
836daad
eacc26a
 
 
 
b8aa7c5
eacc26a
 
 
88d63cf
836daad
eacc26a
57e4050
b8aa7c5
4eada0f
 
 
 
 
 
b8aa7c5
4eada0f
b8aa7c5
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
import gradio as gr
import os
from lumaai import AsyncLumaAI
import asyncio
import aiohttp

async def get_camera_motions():
    api_key = os.getenv("LMGEN_KEY")
    if not api_key:
        raise gr.Error("LMGEN_KEY environment variable is not set.")
    
    client = AsyncLumaAI(auth_token=api_key)
    try:
        motions = await client.generations.camera_motion.list()
        return [str(motion) for motion in motions]  # Convert each motion to a string
    except Exception as e:
        print(f"Error fetching camera motions: {str(e)}")
        return []

async def generate_video(client, prompt, loop=False, aspect_ratio="16:9", camera_motion=None, extend_id=None, reverse_extend_id=None, interpolate_ids=None, progress=gr.Progress()):
    generation_params = {
        "prompt": prompt,
        "loop": loop,
        "aspect_ratio": aspect_ratio
    }

    if camera_motion:
        generation_params["prompt"] += f" {camera_motion}"

    if extend_id:
        generation_params["keyframes"] = {
            "frame0": {
                "type": "generation",
                "id": extend_id
            }
        }
    elif reverse_extend_id:
        generation_params["keyframes"] = {
            "frame1": {
                "type": "generation",
                "id": reverse_extend_id
            }
        }
    elif interpolate_ids:
        generation_params["keyframes"] = {
            "frame0": {
                "type": "generation",
                "id": interpolate_ids[0]
            },
            "frame1": {
                "type": "generation",
                "id": interpolate_ids[1]
            }
        }

    progress(0, desc="Initiating video generation...")
    generation = await client.generations.create(**generation_params)
    
    progress(0.1, desc="Video generation started. Waiting for completion...")
    
    # Poll for completion
    start_time = asyncio.get_event_loop().time()
    while True:
        status = await client.generations.get(id=generation.id)
        if status.state == "completed":
            break
        elif status.state == "failed":
            raise Exception("Video generation failed")
        
        # Update progress based on time elapsed (assuming 60 seconds total)
        elapsed_time = asyncio.get_event_loop().time() - start_time
        progress_value = min(0.1 + (elapsed_time / 60) * 0.8, 0.9)
        progress(progress_value, desc="Generating video...")
        
        await asyncio.sleep(5)

    progress(0.9, desc="Downloading generated video...")
    
    # Download the video
    video_url = status.assets.video
    async with aiohttp.ClientSession() as session:
        async with session.get(video_url) as resp:
            if resp.status == 200:
                file_name = f"luma_ai_generated_{generation.id}.mp4"
                with open(file_name, 'wb') as fd:
                    while True:
                        chunk = await resp.content.read(1024)
                        if not chunk:
                            break
                        fd.write(chunk)
    
    progress(1.0, desc="Video generation complete!")
    return file_name

async def text_to_video(prompt, loop, aspect_ratio, camera_motion, extend_id, reverse_extend_id, interpolate_id1, interpolate_id2, progress=gr.Progress()):
    api_key = os.getenv("LMGEN_KEY")
    if not api_key:
        raise gr.Error("LMGEN_KEY environment variable is not set.")
    
    client = AsyncLumaAI(auth_token=api_key)
    
    try:
        interpolate_ids = None
        if interpolate_id1 and interpolate_id2:
            interpolate_ids = [interpolate_id1, interpolate_id2]

        video_path = await generate_video(
            client, prompt, loop, aspect_ratio, camera_motion, 
            extend_id, reverse_extend_id, interpolate_ids, progress
        )
        return video_path, ""
    except Exception as e:
        return None, f"An error occurred: {str(e)}"

async def image_to_video(prompt, image_url, loop, aspect_ratio, camera_motion, progress=gr.Progress()):
    api_key = os.getenv("LMGEN_KEY")
    if not api_key:
        raise gr.Error("LMGEN_KEY environment variable is not set.")
    
    client = AsyncLumaAI(auth_token=api_key)
    
    try:
        generation_params = {
            "prompt": prompt + (f" {camera_motion}" if camera_motion else ""),
            "loop": loop,
            "aspect_ratio": aspect_ratio,
            "keyframes": {
                "frame0": {
                    "type": "image",
                    "url": image_url
                }
            }
        }
        
        progress(0, desc="Initiating video generation from image...")
        generation = await client.generations.create(**generation_params)
        
        progress(0.1, desc="Video generation started. Waiting for completion...")
        
        # Poll for completion
        start_time = asyncio.get_event_loop().time()
        while True:
            status = await client.generations.get(id=generation.id)
            if status.state == "completed":
                break
            elif status.state == "failed":
                raise Exception("Video generation failed")
            
            # Update progress based on time elapsed (assuming 60 seconds total)
            elapsed_time = asyncio.get_event_loop().time() - start_time
            progress_value = min(0.1 + (elapsed_time / 60) * 0.8, 0.9)
            progress(progress_value, desc="Generating video...")
            
            await asyncio.sleep(5)

        progress(0.9, desc="Downloading generated video...")
        
        # Download the video
        video_url = status.assets.video
        async with aiohttp.ClientSession() as session:
            async with session.get(video_url) as resp:
                if resp.status == 200:
                    file_name = f"luma_ai_generated_{generation.id}.mp4"
                    with open(file_name, 'wb') as fd:
                        while True:
                            chunk = await resp.content.read(1024)
                            if not chunk:
                                break
                            fd.write(chunk)
        
        progress(1.0, desc="Video generation complete!")
        return file_name, ""
    except Exception as e:
        return None, f"An error occurred: {str(e)}"

with gr.Blocks() as demo:
    gr.Markdown("# Luma AI Text-to-Video Demo")
    
    with gr.Tab("Text to Video"):
        prompt = gr.Textbox(label="Prompt")
        generate_btn = gr.Button("Generate Video")
        video_output = gr.Video(label="Generated Video")
        error_output = gr.Textbox(label="Error Messages", visible=True)
        
        with gr.Accordion("Advanced Options", open=False):
            loop = gr.Checkbox(label="Loop", value=False)
            aspect_ratio = gr.Dropdown(label="Aspect Ratio", choices=["16:9", "1:1", "9:16", "4:3", "3:4"], value="16:9")
            camera_motion = gr.Dropdown(label="Camera Motion", choices=[])
            extend_id = gr.Textbox(label="Extend Video ID (optional)")
            reverse_extend_id = gr.Textbox(label="Reverse Extend Video ID (optional)")
            with gr.Row():
                interpolate_id1 = gr.Textbox(label="Interpolate Video ID 1 (optional)")
                interpolate_id2 = gr.Textbox(label="Interpolate Video ID 2 (optional)")
        
        generate_btn.click(
            text_to_video,
            inputs=[prompt, loop, aspect_ratio, camera_motion, extend_id, reverse_extend_id, interpolate_id1, interpolate_id2],
            outputs=[video_output, error_output]
        )
    
    with gr.Tab("Image to Video"):
        img_prompt = gr.Textbox(label="Prompt")
        img_url = gr.Textbox(label="Image URL")
        img_generate_btn = gr.Button("Generate Video from Image")
        img_video_output = gr.Video(label="Generated Video")
        img_error_output = gr.Textbox(label="Error Messages", visible=True)
        
        with gr.Accordion("Advanced Options", open=False):
            img_loop = gr.Checkbox(label="Loop", value=False)
            img_aspect_ratio = gr.Dropdown(label="Aspect Ratio", choices=["16:9", "1:1", "9:16", "4:3", "3:4"], value="16:9")
            img_camera_motion = gr.Dropdown(label="Camera Motion", choices=[])
        
        img_generate_btn.click(
            image_to_video,
            inputs=[img_prompt, img_url, img_loop, img_aspect_ratio, img_camera_motion],
            outputs=[img_video_output, img_error_output]
        )

    async def update_camera_motions():
        try:
            motions = await get_camera_motions()
            return {camera_motion: gr.update(choices=motions), img_camera_motion: gr.update(choices=motions)}
        except Exception as e:
            print(f"Error updating camera motions: {str(e)}")
            return {camera_motion: gr.update(choices=[]), img_camera_motion: gr.update(choices=[])}

    demo.load(update_camera_motions)

demo.queue().launch(debug=True)