File size: 4,740 Bytes
7bd850b | 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 | import os
import json
import numpy as np
import matplotlib.pyplot as plt
from matplotlib import rcParams
from PIL import Image
# 设置全局字体为 Times New Roman
plt.rcParams['font.family'] = 'serif'
# 加载 JSON 数据
json_path = "/Users/gongsitong/Desktop/dataset_processing/Ref_LVOS/video_exp.json"
with open(json_path, 'r') as f:
json_dict = json.load(f)["videos"]
# 获取每个视频的帧数
gt_mask_path = "/Users/gongsitong/Desktop/dataset_processing/Ref_LVOS/Annotations"
occlusion_list = []
change_time_list = []
occlusion_frame_list = []
num_valid_masks = 0
for vid_name, video_dict in json_dict.items():
new_obj_id = 0 # 该视频中实例的数量
object_list = []
for exp_id, exp_dict in video_dict["expressions"].items():
object_list.extend(exp_dict["obj_id"])
# for obj_id in exp_dict["obj_id"]:
# if int(obj_id) != new_obj_id:
# new_obj_id = int(obj_id)
# object_list.append(obj_id)
vid_frames = video_dict["frames"]
object_list = list(set(object_list))
# obj_ids = list(range(num_object))
if "LongVOS" in vid_name:
for obj_id in object_list:
total_frames = len(vid_frames)
num_occlusion_frames = 0
per_obj_change_time_list = []
for frame_idx, frame_name in enumerate(vid_frames):
# processing the gt mask
gt_mask_ = np.array(Image.open(os.path.join(gt_mask_path, vid_name, frame_name + '.png')))
tmp_masks = np.zeros_like(gt_mask_)
tmp_masks[gt_mask_ == int(obj_id)] = 1
gt_mask = tmp_masks
gt_mask[gt_mask == 255] = 0
if np.all(gt_mask == 0):
num_occlusion_frames += 1
per_obj_change_time_list.append(0)
else:
per_obj_change_time_list.append(1)
num_valid_masks += 1
change_time_list.append(int(np.count_nonzero(np.diff(per_obj_change_time_list))))
occlusion_list.append(num_occlusion_frames / total_frames)
occlusion_frame_list.append(num_occlusion_frames)
else:
anno_path = os.path.join(gt_mask_path, vid_name)
mask_list = [os.path.join(anno_path, d) for d in os.listdir(anno_path)]
mask_list.sort()
for obj_id in object_list:
mask_path = mask_list[int(obj_id) - 1]
mask_path_list = sorted(os.listdir(mask_path))
total_frames = len(mask_path_list)
per_obj_change_time_list = []
num_occlusion_frames = 0
for i, mask_path_ in enumerate(mask_path_list):
mask_path_ = os.path.join(mask_path, mask_path_)
mask = np.array(Image.open(mask_path_)) > 0
if np.all(mask == 0) :
num_occlusion_frames += 1
per_obj_change_time_list.append(0)
else:
num_valid_masks += 1
per_obj_change_time_list.append(1)
change_time_list.append(int(np.count_nonzero(np.diff(per_obj_change_time_list))))
occlusion_list.append(num_occlusion_frames / total_frames)
occlusion_frame_list.append(num_occlusion_frames)
print(occlusion_list)
print(change_time_list)
print(occlusion_frame_list)
print("num_valid_masks: {}".format(num_valid_masks))
with open("/Users/gongsitong/Desktop/dataset_processing/Ref_LVOS/occlusion_list.txt", "w", encoding="utf-8") as f:
for item in occlusion_list:
f.write(f"{item}\n")
print("List has been saved to occlusion_list.txt")
with open("/Users/gongsitong/Desktop/dataset_processing/Ref_LVOS/change_time_list.txt", "w", encoding="utf-8") as f:
for item in change_time_list:
f.write(f"{item}\n")
print("List has been saved to change_time_list.txt")
with open("/Users/gongsitong/Desktop/dataset_processing/Ref_LVOS/occlusion_frame_list.txt", "w", encoding="utf-8") as f:
for item in occlusion_frame_list:
f.write(f"{item}\n")
print("List has been saved to occlusion_frame_list.txt")
# bins = np.arange(0, 1.1, 0.1) # 生成 [0.0, 0.1, ..., 1.0]
# # 绘制直方图
# plt.figure(figsize=(6, 4))
# plt.hist(occlusion_list, bins=bins, color=(127 / 255, 255 / 255, 0), edgecolor='black')
# # 设置标签与标题
# plt.xlabel("Occlusion Ratio")
# plt.ylabel("Instance Count")
# plt.title("Occlusion Ratio Distribution (bin=0.1)")
# plt.xticks(bins) # 设置刻度更精确
# plt.tight_layout()
# # 保存图像
# plt.savefig("occlusion_ratio_distribution.png", dpi=300)
# plt.close() |