Spaces:
Sleeping
Sleeping
File size: 12,229 Bytes
463afdd |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 |
from fastapi import FastAPI, UploadFile, File, HTTPException, WebSocket, WebSocketDisconnect
from fastapi.middleware.cors import CORSMiddleware
from fastapi.responses import JSONResponse, StreamingResponse
from pydantic import BaseModel
from typing import Optional, Literal
import asyncio
import time
import hashlib
import io
# Import our utilities
import sys
from pathlib import Path
sys.path.append(str(Path(__file__).parent.parent))
from config import get_settings
from utils.model_loader import ModelManager
from utils.image_processing import (
load_image_from_bytes,
load_image_from_base64,
array_to_base64,
depth_to_colormap,
create_side_by_side
)
from utils.demo_depth import generate_smart_depth
# Initialize FastAPI app
app = FastAPI(
title="Dimensio API",
description="Add Dimension to Everything - High-performance depth estimation and 3D visualization API",
version="1.0.0"
)
settings = get_settings()
# CORS middleware
app.add_middleware(
CORSMiddleware,
allow_origins=settings.CORS_ORIGINS,
allow_credentials=True,
allow_methods=["*"],
allow_headers=["*"],
)
# Global model manager
model_manager = ModelManager()
DEMO_MODE = False # Will be set to True if no models available
# Request/Response models
class DepthRequest(BaseModel):
"""Request model for depth estimation"""
image: str # Base64 encoded image
model: Literal["small", "large"] = "small"
output_format: Literal["grayscale", "colormap", "both"] = "colormap"
colormap: Literal["inferno", "viridis", "plasma", "turbo"] = "inferno"
class DepthResponse(BaseModel):
"""Response model for depth estimation"""
depth_map: str # Base64 encoded depth map
metadata: dict
processing_time_ms: float
# Startup/shutdown events
@app.on_event("startup")
async def startup_event():
"""Initialize models on startup"""
print(">> Starting Dimensio API...")
try:
# Load small model (fast preview)
small_model_path = Path(settings.MODEL_CACHE_DIR) / settings.DEPTH_MODEL_SMALL
if small_model_path.exists():
model_manager.load_model(
"small",
str(small_model_path),
use_gpu=settings.USE_GPU,
use_tensorrt=settings.TRT_OPTIMIZATION
)
print("[+] Small model loaded")
else:
print(f"[!] Small model not found: {small_model_path}")
# Load large model (high quality)
large_model_path = Path(settings.MODEL_CACHE_DIR) / settings.DEPTH_MODEL_LARGE
if large_model_path.exists():
model_manager.load_model(
"large",
str(large_model_path),
use_gpu=settings.USE_GPU,
use_tensorrt=settings.TRT_OPTIMIZATION
)
print("[+] Large model loaded")
else:
print(f"[!] Large model not found: {large_model_path}")
if not model_manager.models:
global DEMO_MODE
DEMO_MODE = True
print("\n[!] No models loaded - Running in DEMO MODE")
print("Demo mode uses synthetic depth maps for testing the UI.")
print("\nTo use real AI models:")
print("1. Run: python download_models.py")
print("2. Place ONNX models in models/cache/")
print("3. Restart the server")
except Exception as e:
print(f"[X] Error loading models: {e}")
print("Server will start but depth estimation will not work.")
@app.on_event("shutdown")
async def shutdown_event():
"""Cleanup on shutdown"""
print(">> Shutting down Depth Flow Pro API...")
# Health check
@app.get("/")
async def root():
"""API health check"""
return {
"name": "Depth Flow Pro API",
"version": "1.0.0",
"status": "online",
"models_loaded": list(model_manager.models.keys())
}
@app.get("/health")
async def health_check():
"""Detailed health check"""
return {
"status": "healthy",
"models": {
name: "loaded" for name in model_manager.models.keys()
},
"gpu_enabled": settings.USE_GPU,
"tensorrt_enabled": settings.TRT_OPTIMIZATION
}
# Depth estimation endpoints
@app.post("/api/v1/depth/preview", response_model=DepthResponse)
async def estimate_depth_preview(file: UploadFile = File(...)):
"""
Fast depth estimation using small model (preview quality)
Optimized for speed, ~50-100ms on GPU
"""
try:
start_time = time.time()
# Load image
image_bytes = await file.read()
image = load_image_from_bytes(image_bytes)
# Check if demo mode or use real model
if DEMO_MODE:
# Use synthetic depth for demo
depth = generate_smart_depth(image)
model_name = "demo"
else:
# Get small model
model = model_manager.get_model("small")
if model is None:
raise HTTPException(
status_code=503,
detail="Small model not loaded. Please check server logs."
)
# Run depth estimation
depth = model.predict(image)
model_name = "small"
# Convert to colormap
depth_colored = depth_to_colormap(depth)
# Encode to base64
depth_base64 = array_to_base64(depth_colored, format='PNG')
processing_time = (time.time() - start_time) * 1000
return DepthResponse(
depth_map=depth_base64,
metadata={
"model": model_name,
"input_size": image.shape[:2],
"output_size": depth.shape[:2],
"demo_mode": DEMO_MODE
},
processing_time_ms=round(processing_time, 2)
)
except Exception as e:
print(f"❌ Error: {type(e).__name__}: {str(e)}")
import traceback
traceback.print_exc()
raise HTTPException(status_code=500, detail=str(e))
@app.post("/api/v1/depth/hq", response_model=DepthResponse)
async def estimate_depth_hq(file: UploadFile = File(...)):
"""
High-quality depth estimation using large model
Slower but more accurate, ~500-1500ms on GPU
"""
try:
start_time = time.time()
# Load image
image_bytes = await file.read()
image = load_image_from_bytes(image_bytes)
# Check if demo mode or use real model
if DEMO_MODE:
# Use synthetic depth for demo
depth = generate_smart_depth(image)
model_name = "demo (HQ)"
else:
# Get large model
model = model_manager.get_model("large")
if model is None:
# Fallback to small model if large not available
model = model_manager.get_model("small")
if model is None:
raise HTTPException(
status_code=503,
detail="No models loaded. Please check server logs."
)
model_name = "small (fallback)"
else:
model_name = "large"
# Run depth estimation
depth = model.predict(image)
# Convert to colormap
depth_colored = depth_to_colormap(depth)
# Encode to base64
depth_base64 = array_to_base64(depth_colored, format='PNG')
processing_time = (time.time() - start_time) * 1000
return DepthResponse(
depth_map=depth_base64,
metadata={
"model": model_name,
"input_size": image.shape[:2],
"output_size": depth.shape[:2],
"demo_mode": DEMO_MODE
},
processing_time_ms=round(processing_time, 2)
)
except Exception as e:
print(f"❌ Error: {type(e).__name__}: {str(e)}")
import traceback
traceback.print_exc()
raise HTTPException(status_code=500, detail=str(e))
@app.post("/api/v1/depth/estimate")
async def estimate_depth(request: DepthRequest):
"""
Depth estimation with custom options
Accepts base64 encoded image
"""
try:
start_time = time.time()
# Load image from base64
image = load_image_from_base64(request.image)
# Get model
model = model_manager.get_model(request.model)
if model is None:
raise HTTPException(
status_code=503,
detail=f"Model '{request.model}' not loaded"
)
# Run depth estimation
depth = model.predict(image)
# Process output based on format
if request.output_format == "grayscale":
output = (depth * 255).astype('uint8')
depth_base64 = array_to_base64(output, format='PNG')
elif request.output_format == "colormap":
import cv2
colormap_dict = {
"inferno": cv2.COLORMAP_INFERNO,
"viridis": cv2.COLORMAP_VIRIDIS,
"plasma": cv2.COLORMAP_PLASMA,
"turbo": cv2.COLORMAP_TURBO
}
depth_colored = depth_to_colormap(depth, colormap_dict[request.colormap])
depth_base64 = array_to_base64(depth_colored, format='PNG')
else: # both
side_by_side = create_side_by_side(image, depth, colormap=True)
depth_base64 = array_to_base64(side_by_side, format='PNG')
processing_time = (time.time() - start_time) * 1000
return DepthResponse(
depth_map=depth_base64,
metadata={
"model": request.model,
"output_format": request.output_format,
"colormap": request.colormap,
"input_size": image.shape[:2],
"output_size": depth.shape[:2]
},
processing_time_ms=round(processing_time, 2)
)
except Exception as e:
print(f"❌ Error: {type(e).__name__}: {str(e)}")
import traceback
traceback.print_exc()
raise HTTPException(status_code=500, detail=str(e))
# WebSocket for streaming
@app.websocket("/api/v1/stream")
async def websocket_endpoint(websocket: WebSocket):
"""
WebSocket endpoint for real-time depth estimation
Supports streaming multiple images
"""
await websocket.accept()
try:
while True:
# Receive image data
data = await websocket.receive_json()
if data.get("action") == "estimate":
start_time = time.time()
# Load image
image = load_image_from_base64(data["image"])
# Get model
model_name = data.get("model", "small")
model = model_manager.get_model(model_name)
if model is None:
await websocket.send_json({
"error": f"Model '{model_name}' not loaded"
})
continue
# Send progress update
await websocket.send_json({
"status": "processing",
"progress": 50
})
# Run depth estimation
depth = model.predict(image)
# Convert to colormap
depth_colored = depth_to_colormap(depth)
depth_base64 = array_to_base64(depth_colored, format='PNG')
processing_time = (time.time() - start_time) * 1000
# Send result
await websocket.send_json({
"status": "complete",
"depth_map": depth_base64,
"processing_time_ms": round(processing_time, 2)
})
except WebSocketDisconnect:
print("WebSocket disconnected")
except Exception as e:
print(f"WebSocket error: {e}")
await websocket.send_json({"error": str(e)})
if __name__ == "__main__":
import uvicorn
uvicorn.run(
"main:app",
host=settings.HOST,
port=settings.PORT,
reload=settings.DEBUG
)
|