File size: 3,640 Bytes
b27cd24 | 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 | import json
import os
from time import sleep
from tqdm import tqdm
from PIL import Image
from google import genai
from google.genai import types
from pydantic import BaseModel
# =========================
# Gemini setup
# =========================
GEMINI_API_KEY = "AIzaSyAfnBWMguUci9GyzW-gBxrxCfOmMExiDnA"
client = genai.Client(api_key=GEMINI_API_KEY)
MODEL_ID = "gemini-3-flash-preview" # or newer if available
# =========================
# JSON schema (STRICT)
# =========================
class QueueAnswer(BaseModel):
number_of_people: int
line_direction: str
end_visible: str
end_location: str
end_camera_direction: str
end_person_description: str
start_visible: str
start_location: str
start_camera_direction: str
start_person_description: str
# =========================
# Prompt (same as GPT version)
# =========================
def build_prompt():
return (
"You are an expert at analyzing a single image of a line of people.\n"
"Return STRICT JSON only.\n\n"
"Fields:\n"
'number_of_people: integer\n'
'line_direction: ["towards", "away", "sideways-left", "sideways-right"]\n'
'end_visible: ["yes","no"]\n'
'end_location: ["far left","center left","center","center right","far right","N/A"]\n'
'end_camera_direction: ["left","right","back","N/A"]\n'
'end_person_description: string\n'
'start_visible: ["yes","no"]\n'
'start_location: ["far left","center left","center","center right","far right","N/A"]\n'
'start_camera_direction: ["left","right","back","N/A"]\n'
'start_person_description: string\n'
)
# =========================
# Single image inference
# =========================
def analyze_image(img_path):
image = Image.open(img_path)
image.thumbnail([512, 512]) # Gemini requirement
response = client.models.generate_content(
model=MODEL_ID,
contents=[
"Return ONLY JSON.",
build_prompt(),
image
],
config=types.GenerateContentConfig(
response_mime_type="application/json",
response_schema=QueueAnswer,
temperature=0.2,
),
)
return response.text
# =========================
# Batch processing
# =========================
def generate_reranking(image_paths, output_file):
with open(output_file, "a", encoding="utf-8") as f:
for img_path in tqdm(image_paths):
basename = os.path.basename(img_path)
try:
result = analyze_image(img_path)
try:
parsed = json.loads(result)
except:
parsed = {"error": "invalid_json", "raw": result}
f.write(basename + "\n")
f.write(json.dumps(parsed) + "\n\n")
f.flush()
except Exception as e:
print(f"Error: {img_path} -> {e}")
f.write(basename + "\n")
f.write(json.dumps({"error": str(e)}) + "\n\n")
sleep(0.2)
# =========================
# Load images (same as before)
# =========================
root = "/scratch/ds5725/linefinder/LineFinder/Images"
subfolders = ["QueuesOutdoors","QueuesInSupermarketNew","QueuesInThemeParks"]
all_files = []
for sub in subfolders:
folder_path = os.path.join(root, sub)
for dirpath, _, filenames in os.walk(folder_path):
for f in filenames:
all_files.append(os.path.abspath(os.path.join(dirpath, f)))
all_files.sort()
generate_reranking(all_files, "gemini_line_luna_olivia.jsonl") |