from fastapi import FastAPI, UploadFile, File
import cv2
import numpy as np
import mediapipe as mp

app = FastAPI(title="Human Anthropometry API")

mp_pose = mp.solutions.pose
pose = mp_pose.Pose(static_image_mode=False)

# ---- Utility ----
def dist(a, b):
    return np.linalg.norm(a - b)

def analyze_frame(image):
    h, w, _ = image.shape
    rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    res = pose.process(rgb)

    if not res.pose_landmarks:
        return None

    lm = res.pose_landmarks.landmark

    def p(i):
        return np.array([lm[i].x * w, lm[i].y * h])

    head = p(0)
    l_sh = p(11)
    r_sh = p(12)
    l_hip = p(23)
    r_hip = p(24)
    l_heel = p(29)
    r_heel = p(30)

    shoulder_px = dist(l_sh, r_sh)
    torso_px = dist((l_sh + r_sh)/2, (l_hip + r_hip)/2)
    height_px = dist(head, (l_heel + r_heel)/2)

    # ---- INDUSTRIAL NORMALIZATION ----
    # Average adult shoulder width ≈ 0.23 × height
    height_cm = (shoulder_px / height_px) * (1 / 0.23) * 100

    shoulder_cm = height_cm * 0.23
    torso_ratio = torso_px / height_px

    # Weight estimation via body volume proxy (not BMI)
    weight_kg = round((height_cm * shoulder_cm * torso_ratio) / 1000, 1)

    confidence = round(min(0.95, 0.6 + torso_ratio), 2)

    return {
        "height_cm": round(height_cm, 1),
        "shoulder_cm": round(shoulder_cm, 1),
        "weight_kg": weight_kg,
        "confidence": confidence
    }

# ---- API ----
@app.post("/analyze")
async def analyze(file: UploadFile = File(...)):
    data = await file.read()
    img = cv2.imdecode(np.frombuffer(data, np.uint8), cv2.IMREAD_COLOR)

    result = analyze_frame(img)
    if not result:
        return {"error": "Human not detected"}

    return result