mbench/gpt_ref-ytvos_numbered_cy.py

import os
import sys
sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), '..')))
import time

from os import path as osp
from io import BytesIO

from mbench.ytvos_ref import build as build_ytvos_ref
import argparse
import opts

import sys
from pathlib import Path
import os
from os import path as osp
import skimage
from io import BytesIO

import numpy as np
import pandas as pd
import regex as re
import json

import cv2
from PIL import Image, ImageDraw
import torch
from torchvision.transforms import functional as F

from skimage import measure                        # (pip install scikit-image)
from shapely.geometry import Polygon, MultiPolygon # (pip install Shapely)

import matplotlib.pyplot as plt
import matplotlib.patches as patches
from matplotlib.collections import PatchCollection
from matplotlib.patches import Rectangle
import textwrap


import ipywidgets as widgets
from IPython.display import display, clear_output

from openai import OpenAI
import base64
import json

def number_objects_and_encode(idx, color_mask=False):
    encoded_frames = {}
    contoured_frames = {}  # New dictionary for original images
    vid_cat_cnts = {}

    vid_meta = metas[idx]
    vid_data = train_dataset[idx]
    vid_id = vid_meta['video']
    frame_indx = vid_meta['sample_indx']
    cat_names = set(vid_meta['obj_id_cat'].values())
    imgs = vid_data[0]
    
    for cat in cat_names:
        cat_frames = []
        contour_frames = []
        frame_cat_cnts = {}

        for i in range(imgs.size(0)):
            frame_name = frame_indx[i]
            frame = np.copy(imgs[i].permute(1, 2, 0).numpy())  
            frame_for_contour = np.copy(imgs[i].permute(1, 2, 0).numpy()) 

            frame_data = vid_data[2][frame_name]
            obj_ids = list(frame_data.keys())

            cat_cnt = 0

            for j in range(len(obj_ids)):
                obj_id = obj_ids[j]
                obj_data = frame_data[obj_id]
                obj_bbox = obj_data['bbox']
                obj_valid = obj_data['valid']
                obj_mask = obj_data['mask'].numpy().astype(np.uint8)
                obj_cat = obj_data['category_name']

                if obj_cat == cat and obj_valid:
                    cat_cnt += 1
                    
                    if color_mask == False:
                        contours, _ = cv2.findContours(obj_mask, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
                        cv2.drawContours(frame, contours, -1, colors[j], 3)
                        for i, contour in enumerate(contours):
                            # 윤곽선 중심 계산
                            moments = cv2.moments(contour)
                            if moments["m00"] != 0:  # 중심 계산 가능 여부 확인
                                cx = int(moments["m10"] / moments["m00"])
                                cy = int(moments["m01"] / moments["m00"])
                            else:
                                cx, cy = contour[0][0]  # 중심 계산 불가시 대체 좌표 사용
                            
                            # 텍스트 배경 (검은색 배경 만들기)
                            font = cv2.FONT_HERSHEY_SIMPLEX
                            text = obj_id
                            text_size = cv2.getTextSize(text, font, 1, 2)[0]
                            text_w, text_h = text_size
                            
                            # 텍스트 배경 그리기 (검은색 배경)
                            cv2.rectangle(frame, (cx - text_w // 2 - 5, cy - text_h // 2 - 5),
                                        (cx + text_w // 2 + 5, cy + text_h // 2 + 5), (0, 0, 0), -1)
                            
                            # 텍스트 그리기 (흰색 텍스트)
                            cv2.putText(frame, text, (cx - text_w // 2, cy + text_h // 2),
                                        font, 1, (255, 255, 255), 2)

                    else:
                        alpha = 0.08

                        colored_obj_mask = np.zeros_like(frame)  
                        colored_obj_mask[obj_mask == 1] = colors[j]
                        frame[obj_mask == 1] = (
                            (1 - alpha) * frame[obj_mask == 1]
                            + alpha * colored_obj_mask[obj_mask == 1]
                        )


                        contours, _ = cv2.findContours(obj_mask, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
                        cv2.drawContours(frame, contours, -1, colors[j], 2)
                        cv2.drawContours(frame_for_contour, contours, -1, colors[j], 2)
                        
                        
                        
                        if len(contours) > 0:
                            largest_contour = max(contours, key=cv2.contourArea)
                            M = cv2.moments(largest_contour)
                            if M["m00"] != 0:  
                                center_x = int(M["m10"] / M["m00"])
                                center_y = int(M["m01"] / M["m00"])
                            else:
                                center_x, center_y = 0, 0

                        font = cv2.FONT_HERSHEY_SIMPLEX
                        text = obj_id

                        font_scale = 0.9  
                        text_size = cv2.getTextSize(text, font, font_scale, 2)[0]                        
                        text_x = center_x - text_size[0] // 1  # 텍스트의 가로 중심
                        text_y = center_y
                        # text_y = center_y + text_size[1] // 2 # 텍스트의 세로 중심

                        # 텍스트 배경 사각형 좌표 계산
                        rect_start = (text_x - 5, text_y - text_size[1] - 5)  # 배경 사각형 좌상단
                        # rect_end = (text_x + text_size[0] + 5, text_y + 5) 
                        rect_end = (text_x + text_size[0] + 5, text_y)

                        cv2.rectangle(frame, rect_start, rect_end, (0, 0, 0), -1)
                        cv2.putText(frame, text, (text_x, text_y), font, 1, (255, 255, 255), 2)

            # plt.figure(figsize=(12, 8))
            # plt.imshow(frame)
            # plt.title(f"frame {frame_name}")
            # plt.tight_layout()
            # plt.axis('off')
            # plt.show()
        
            buffer = BytesIO()
            frame = Image.fromarray(frame)
            frame.save(buffer, format='jpeg')
            buffer.seek(0)
            cat_frames.append(base64.b64encode(buffer.read()).decode("utf-8"))
            frame_cat_cnts[frame_name] = cat_cnt

            buffer.seek(0)  # Reuse buffer instead of creating a new one
            buffer.truncate()
            frame_for_contour = Image.fromarray(frame_for_contour)
            frame_for_contour.save(buffer, format='jpeg')
            buffer.seek(0)
            contour_frames.append(base64.b64encode(buffer.read()).decode("utf-8"))
        
        encoded_frames[cat] = cat_frames
        contoured_frames[cat] = contour_frames
        vid_cat_cnts[cat] = frame_cat_cnts
    
    return encoded_frames, vid_cat_cnts, contoured_frames 


def getCaption(idx, model='gpt-4o', color_mask=True):
    vid_meta = metas[idx]
    vid_data = train_dataset[idx]
    vid_id = vid_meta['video']
    print(f"vid id: {vid_id}\n")

    frame_indx = vid_meta['sample_indx']             # e.g. [4, 7, 9, 16]
    cat_names = set(vid_meta['obj_id_cat'].values()) # e.g. {"person", "elephant", ...}
    all_captions = dict()

    base64_frames, vid_cat_cnts, contoured_frames = number_objects_and_encode(idx, color_mask)
    #marked = "mask with boundary" if color_mask else "boundary"

    for cat_name in list(cat_names) :

        is_movable = False  
        if cat_name in ytvos_category_valid_list :
            is_movable = True
            
        if not is_movable:
            print(f"Skipping {cat_name}: Determined to be non-movable.", end='\n\n')


        image_captions = {}
        captioner = OpenAI()
        cat_base64_frames = base64_frames[cat_name]
        cont_base64_frames = contoured_frames[cat_name]

        for i in range(len(cat_base64_frames)):
            frame_name = frame_indx[i]
            cont_base64_image = cont_base64_frames[i]
            base64_image = cat_base64_frames[i]
            should_filter = False
            frame_cat_cnts = vid_cat_cnts[cat_name][frame_name]

            if frame_cat_cnts >= 2:
                should_filter = True
            else:
                print(f"Skipping {cat_name}: There is single or no object.", end='\n\n')

            if is_movable and should_filter:
                #1단계: 필터링
                print(f"-----------category name: {cat_name}, frame name: {frame_name}")
                caption_filter_text = f"""
                You are a visual assistant analyzing a single frame from a video.
                In this frame, I have labeled {frame_cat_cnts} {cat_name}(s), each with a bright numeric ID at its center and a visible marker.

                Are {cat_name}s in the image performing all different and recognizable actions or postures? 
                Consider differences in body pose (standing, sitting, holding hands up, grabbing object, facing the camera, stretching, walking...), motion cues (inferred from the momentary stance or position), 
                facial expressions, and any notable interactions with objects or other {cat_name}s or people. 

                Only focus on obvious, prominent actions that can be reliably identified from this single frame.
     
                - Respond with "YES" if:
                1) Most of {cat_name}s exhibit clearly different, unique actions or poses.
                    (e.g. standing, sitting, bending, stretching, showing its back, or turning toward the camera.)
                2) You can see visible significant differences in action and posture, that an observer can identify at a glance.
                3) Interaction Variability: Each {cat_name} is engaged in a different type of action, such as one grasping an object while another is observing.

                - Respond with "NONE" if:
                1) The actions or pose are not clearly differentiable or too similar.
                2) Minimal or Ambiguous Motion: The frame does not provide clear evidence of distinct movement beyond subtle shifts in stance.
                3) Passive or Neutral Poses: If multiple {cat_name}(s) are simply standing or sitting without an obvious difference in orientation or motion

                Answer strictly with either "YES" or "NONE".
                """
                
                response1 = captioner.chat.completions.create(
                    model=model,
                    messages=[
                        {
                            "role": "user",
                            "content": [
                                {                            
                                    "type": "text",
                                    "text": caption_filter_text,
                                },
                                {
                                    "type": "image_url",
                                    "image_url": {"url": f"data:image/jpeg;base64,{base64_image}"},
                                }
                            ],
                        }
                    ],
                )
                response_content = response1.choices[0].message.content
                should_caption = True if "yes" in response_content.lower() else False
                print(f"are {cat_name}s distinguished by action: {response_content}", end='\n\n')
                
            else:
                should_caption = False

            #2단계: dense caption 만들기
            dense_caption_prompt_1 =  f"""You are a visual assistant that can analyze a single frame of a video and create referring expressions for each object.
                                        In the given frame, I labeled {frame_cat_cnts} {cat_name}s by marking each with a bright numeric ID at the center and its boundary.
                                        I want to use your expressions to create a action-centric referring expression dataset.
                                        Therefore, your expressions for these {cat_name}s should describe unique action of each object. 
                                        
                                        1. Focus only on clear, unique, and prominent actions that distinguish each object.
                                        2. Avoid describing actions that are too minor, ambiguous, or not visible from the image.
                                        3. Avoid subjective terms such as 'skilled', 'controlled', or 'focused'. Only describe observable actions.
                                        4. Do not include common-sense or overly general descriptions like 'the elephant walks'.
                                        5. Use dynamic action verbs (e.g., holding, throwing, jumping, inspecting) to describe interactions, poses, or movements.
                                        6. Avoid overly detailed or speculative descriptions such as 'slightly moving its mouth' or 'appears to be anticipating'.
                                        7. Pretend you are observing the scene directly, avoiding phrases like 'it seems' or 'based on the description'.
                                        8. Include interactions with objects or other entities when they are prominent and observable.
                                        9. If the image contains multiple {cat_name}s, describe the actions of each individually and ensure the descriptions are non-overlapping and specific.
                                        10. Do not include descriptions of appearance such as clothes, color, size, shape etc.
                                        11. Do not include relative position between objects such as 'the left elephant' because left/right can be ambiguous.
                                        12. Do not mention object IDs.
                                        13. Use '{cat_name}' as the noun for the referring expressions.

                                        Keep in mind that you should not group the objects, e.g., 2-5. people: xxx, be sure to describe each object separately (one by one). 
                                        Output referring expressions for each object id.
                                        """
                                        
            dense_caption_prompt = f"""
            You are a visual assistant analyzing a single frame of a video.  
            In the given frame, I labeled {frame_cat_cnts} {cat_name}s by marking each with a bright numeric ID at the center and its boundary.

            I want to use your expressions to create an **action-centric referring expression** dataset.  
            Please describe each {cat_name} using **clearly observable** and **specific** actions.

            ---
            ## Guidelines:
            1. **Focus on visible, prominent actions** only (e.g., running, pushing, grasping an object).  
            2. **Avoid describing minor or ambiguous actions** (e.g., "slightly moving a paw", "slightly tilting head").  
            3. **Do not include subjective or speculative descriptions** (e.g., “it seems excited” or “it might be preparing to jump”).  
            4. **Avoid vague expressions** like "interacting with something" or "engaging with another object." Instead, specify the action (e.g., "grabbing a stick," "pressing a button").  
            5. **Use dynamic action verbs** (holding, throwing, inspecting, leaning, pressing) to highlight body movement or object/animal interaction.  
            6. If multiple {cat_name}s appear, ensure each description **differentiates** their actions.  
            7. Base your description on these action definitions:
            - Avoid using term 'minimal' or 'slightly'.
            - General body movement, body position, or pattern which is prominent. (e.g. "lifting head up", "facing towards", "showing its back")
            - details such as motion and intention, facial with object manipulation
            - movements with objects or other entities when they are prominent and observable. expression should be specific.
                (e.g., "pushing another person" (O), "engaging with someone" (X) "interacting with another person" (X))           
            ---

            ## Output Format:
            - For each labeled {cat_name}, output **exactly one line**. Your answer should contain details and follow the following format :
                object id. using {cat_name} as subject noun, action-oriented description        
                (e.g. 1. the person is holding ski poles and skiing on a snow mountain, with his two legs bent forward.)
            - **Only include the currently labeled category** in each line (e.g., if it’s a person, do not suddenly label it as other object/animal).

            ### Example
            If the frame has 2 labeled bears, your output should look like:
            1. the bear reaching his right arm while leaning forward to capture the prey
            2. a bear standing upright facing right, touching the bike aside

            ---
            **Do not include** appearance details (e.g., color, size, texture) or relative positioning (e.g., “on the left/right”).
            **Do not include object IDs** or reference them (e.g., "Person 1" or "object 2" is not allowed).
            **Do not include markdown** in the output.
            Keep in mind that you should not group the objects, e.g., 2-5. people: xxx, be sure to describe each object separately (one by one). 
            For each labeled {cat_name}, output referring expressions for each object id.
            """
            MAX_RETRIES = 2
            retry_count = 0

            if should_caption:
                while retry_count < MAX_RETRIES:

                    response2 = captioner.chat.completions.create(
                        model=model,
                        messages=[
                            {
                                "role": "user",
                                "content": [
                                    {
                                        "type": "text",
                                        "text": dense_caption_prompt,
                                    },
                                    {
                                        "type": "image_url",
                                        "image_url": {"url": f"data:image/jpeg;base64,{base64_image}"},
                                    },
                                ],
                            }
                        ],
                    )

                    # caption = response2.choices[0].message.content
                    #print(f"{image_path} - {frame_name}: {caption}")

                    caption = response2.choices[0].message.content.strip() 
                    caption_lower = caption.lower().lstrip() 

                    if caption_lower.startswith("1.") and not any(
                        phrase in caption_lower for phrase in ["i'm sorry", "please", "can't help"]
                    ):
                        break

                    print(f"Retrying caption generation... ({retry_count + 1}/{MAX_RETRIES})")
                    retry_count += 1
                    time.sleep(2) 

                if retry_count == MAX_RETRIES:
                    caption = None
                    print("Max retries reached. Caption generation failed.")
                    
            else:
                caption = None

            image_captions[frame_name] = caption
        all_captions[cat_name] = image_captions
        
    # final : also prepare valid object ids
    valid_obj_ids = dict()

    for cat in cat_names:
        if cat in ytvos_category_valid_list:
            obj_id_cat = vid_meta['obj_id_cat']
            valid_cat_ids = []
            for obj_id in list(obj_id_cat.keys()):
                if obj_id_cat[obj_id] == cat:
                    valid_cat_ids.append(obj_id)
            valid_obj_ids[cat] = valid_cat_ids

    return vid_id, all_captions, valid_obj_ids



if __name__ == '__main__':
    parser = argparse.ArgumentParser('ReferFormer training and evaluation script', parents=[opts.get_args_parser()])
    parser.add_argument('--save_caption_path', type=str, default="mbench/numbered_captions.json")
    parser.add_argument('--save_valid_obj_ids_path', type=str, default="mbench/numbered_valid_obj_ids.json")
    
    args = parser.parse_args()

    #==================데이터 불러오기===================
    # 전체 데이터셋
    train_dataset = build_ytvos_ref(image_set = 'train', args = args)

    # 전체 데이터셋 메타데이터
    metas = train_dataset.metas

    # 색상 후보 8개 (RGB 형식)
    colors = [
        (255, 0, 0),    # Red
        (0, 255, 0),    # Green
        (0, 0, 255),    # Blue
        (255, 255, 0),  # Yellow
        (255, 0, 255),  # Magenta
        (0, 255, 255),  # Cyan
        (128, 0, 128),  # Purple
        (255, 165, 0)   # Orange
    ]

    ytvos_category_valid_list = [
    'airplane', 'ape', 'bear', 'bird', 'boat', 'bus', 'camel', 'cat', 'cow', 'crocodile', 
    'deer', 'dog', 'dolphin', 'duck', 'eagle', 'earless_seal', 'elephant', 'fish', 'fox', 'frog', 
    'giant_panda', 'giraffe', 'hedgehog', 'horse', 'leopard', 'lion', 'lizard', 
    'monkey', 'motorbike', 'mouse', 'owl', 'parrot', 'penguin', 'person', 
    'rabbit', 'raccoon', 'sedan', 'shark', 'sheep', 'snail', 'snake',  
    'squirrel', 'tiger', 'train', 'truck', 'turtle', 'whale', 'zebra'
    ]

    #==================gpt 돌리기===================
    os.environ['OPENAI_API_KEY'] = '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'
    
    result_captions = {}
    result_valid_obj_ids = {}

    for i in range(370):
        vid_id, all_captions, valid_obj_ids = getCaption(i, color_mask=False)

        if vid_id not in result_captions:
            result_captions[vid_id] = all_captions
        if vid_id not in result_valid_obj_ids:
            result_valid_obj_ids[vid_id] = valid_obj_ids

    print("Finished!", flush=True)

    with open(args.save_caption_path, "w") as file:
        json.dump(result_captions, file, indent=4)
    
    with open(args.save_valid_obj_ids_path, "w") as file:
        json.dump(result_valid_obj_ids, file, indent=4)