scripts/vis_handal_gt.py

import argparse
import json
import tqdm
import cv2
import os
import numpy as np
import random
from pycocotools.mask import encode, decode, frPyObjects

EVALMODE = "test"


def fuse_mask(mask_list):
    fused_mask = np.zeros_like(mask_list[0])
    for mask in mask_list:
        fused_mask[mask == 1] = 1
    return fused_mask


def blend_mask(input_img, binary_mask, alpha=0.5, color="g"):
    if input_img.ndim == 2:
        return input_img
    mask_image = np.zeros(input_img.shape, np.uint8)
    if color == "r":
        mask_image[:, :, 0] = 255
    if color == "g":
        mask_image[:, :, 1] = 255
    if color == "b":
        mask_image[:, :, 2] = 255
    if color == "o":
        mask_image[:, :, 0] = 255
        mask_image[:, :, 1] = 165
        mask_image[:, :, 2] = 0
    if color == "c":
        mask_image[:, :, 0] = 0
        mask_image[:, :, 1] = 255
        mask_image[:, :, 2] = 255
    if color == "p":
        mask_image[:, :, 0] = 128
        mask_image[:, :, 1] = 0
        mask_image[:, :, 2] = 128


    mask_image = mask_image * np.repeat(binary_mask[:, :, np.newaxis], 3, axis=2)
    blend_image = input_img[:, :, :].copy()
    pos_idx = binary_mask > 0
    for ind in range(input_img.ndim):
        ch_img1 = input_img[:, :, ind]
        ch_img2 = mask_image[:, :, ind]
        ch_img3 = blend_image[:, :, ind]
        ch_img3[pos_idx] = alpha * ch_img1[pos_idx] + (1 - alpha) * ch_img2[pos_idx]
        blend_image[:, :, ind] = ch_img3
    return blend_image


def upsample_mask(mask, frame):
    H, W = frame.shape[:2]
    mH, mW = mask.shape[:2]

    if W > H:
        ratio = mW / W
        h = H * ratio
        diff = int((mH - h) // 2)
        if diff == 0:
            mask = mask
        else:
            mask = mask[diff:-diff]
    else:
        ratio = mH / H
        w = W * ratio
        diff = int((mW - w) // 2)
        if diff == 0:
            mask = mask
        else:
            mask = mask[:, diff:-diff]

    mask = cv2.resize(mask, (W, H))
    return mask


def downsample(mask, frame):
    H, W = frame.shape[:2]
    mH, mW = mask.shape[:2]

    mask = cv2.resize(mask, (W, H))
    return mask


if __name__ == "__main__":
   

    color = ['g', 'r', 'b', 'o', 'c', 'p']
    data_path = "/data/work-gcp-europe-west4-a/yuqian_fu/datasets/HANDAL/handal_dataset_whisks/train"
    json_path = "/data/work-gcp-europe-west4-a/yuqian_fu/datasets/HANDAL/handal_dataset_whisks/handal_dataset_whisks_test100.json" #debug
    output_path = "/data/work-gcp-europe-west4-a/yuqian_fu/datasets/HANDAL/handal_dataset_whisks/vis"
    mask_path = "/data/work-gcp-europe-west4-a/yuqian_fu/datasets/HANDAL/handal_dataset_whisks/mask_predictions/train"

    with open(json_path, "r") as fp:
        datas = json.load(fp)

    takes_ids = os.listdir(mask_path)
    for take_id in tqdm.tqdm(takes_ids):
        data_list = []
        for data in datas:
            if data["video_name"] == take_id:
                data_list.append(data)
        
        #开始按帧保存fuse-mask
        for data in data_list:
            name = data["image"].split("/")[-1]
            frame_idx = name.split(".")[0]
            #target gt
            frame_target = cv2.imread(
                            f"{data_path}/{data['image']}"
                        )

            for i,ann in enumerate(data["anns"]):
                mask = decode(ann["segmentation"])
                mask = downsample(mask, frame_target)
                out = blend_mask(frame_target, mask, color=color[0])
                os.makedirs(
                            f"{output_path}/{take_id}/obj_{i}/target_gt", #debug
                            exist_ok=True,
                        )
                cv2.imwrite(
                            f"{output_path}/{take_id}/obj_{i}/target_gt/{frame_idx}.jpg",  #debug
                            out,
                        )
        
            
            #query gt
            frame_query = cv2.imread(
                            f"{data_path}/{data['first_frame_image']}"
                        )


            for i,ann in enumerate(data["first_frame_anns"]):
                mask = decode(ann["segmentation"])
                mask = downsample(mask, frame_query)
                out = blend_mask(frame_query, mask, color=color[0])
                os.makedirs(
                            f"{output_path}/{take_id}/obj_{i}/query_gt",  #debug
                            exist_ok=True,
                        )
                cv2.imwrite(
                            f"{output_path}/{take_id}/obj_{i}/query_gt/{frame_idx}.jpg",  #debug
                            out,
                        )