File size: 7,208 Bytes
77731f3 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 | #!/usr/bin/env python3
"""
HM3D语义信息读取工具
用于从semantic.txt或semantic.glb文件中读取房间语义信息
"""
import os
from pathlib import Path
from typing import Dict, List, Optional, Tuple
import numpy as np
def parse_semantic_txt(semantic_txt_path: str) -> Dict[int, List[int]]:
"""
解析semantic.txt文件，提取房间ID到对象ID的映射
semantic.txt格式：
ID,颜色,类别名,房间ID
Args:
semantic_txt_path: semantic.txt文件路径
Returns:
room_to_objects: {room_id: [object_ids]} 字典
"""
room_to_objects: Dict[int, List[int]] = {}
if not os.path.exists(semantic_txt_path):
return room_to_objects
with open(semantic_txt_path, 'r', encoding='utf-8') as f:
lines = f.readlines()
# 跳过第一行标题（如果有）
start_idx = 0
if len(lines) > 0 and 'HM3D Semantic Annotations' in lines[0]:
start_idx = 1
for line in lines[start_idx:]:
line = line.strip()
if not line:
continue
# 解析格式: ID,颜色,类别名,房间ID
parts = line.split(',')
if len(parts) >= 4:
try:
object_id = int(parts[0].strip())
room_id_str = parts[3].strip()
# 房间ID可能是数字或空字符串
if room_id_str.isdigit():
room_id = int(room_id_str)
if room_id not in room_to_objects:
room_to_objects[room_id] = []
room_to_objects[room_id].append(object_id)
except (ValueError, IndexError):
continue
return room_to_objects
def load_semantic_from_glb(semantic_glb_path: str) -> Dict[int, List[int]]:
"""
从semantic.glb文件中读取语义信息
Args:
semantic_glb_path: semantic.glb文件路径
Returns:
room_to_objects: {room_id: [object_ids]} 字典
"""
room_to_objects: Dict[int, List[int]] = {}
if not os.path.exists(semantic_glb_path):
return room_to_objects
try:
import trimesh
scene = trimesh.load(semantic_glb_path, process=False)
if isinstance(scene, trimesh.Scene):
# 遍历场景中的所有几何体
for name, geom in scene.geometry.items():
if isinstance(geom, trimesh.Trimesh):
# 尝试从对象名称或材质中提取房间ID
# HM3D的semantic.glb中，对象名称可能包含房间信息
# 或者通过材质索引来区分房间
# 这里需要根据实际格式调整
# 暂时返回空字典，因为semantic.txt更可靠
pass
except ImportError:
print("[WARN] trimesh未安装，无法从GLB读取语义信息")
except Exception as e:
print(f"[WARN] 从GLB读取语义信息失败: {e}")
return room_to_objects
def find_semantic_file(mesh_path: str) -> Optional[str]:
"""
根据GLB文件路径查找对应的语义文件
查找顺序：
1. semantic.txt文件（优先）
2. semantic.glb文件
Args:
mesh_path: GLB文件路径
Returns:
语义文件路径，如果不存在则返回None
"""
mesh_path_obj = Path(mesh_path)
mesh_dir = mesh_path_obj.parent
mesh_stem = mesh_path_obj.stem # 不含扩展名的文件名
# 尝试查找semantic.txt
# 路径模式：从 hm3d-example-glb-v0.2/... 到 hm3d-example-semantic-annots-v0.2/...
semantic_txt_path = None
semantic_glb_path = None
# 方法1: 在同一目录下查找
semantic_txt_candidate = mesh_dir / f"{mesh_stem}.semantic.txt"
if semantic_txt_candidate.exists():
semantic_txt_path = str(semantic_txt_candidate)
semantic_glb_candidate = mesh_dir / f"{mesh_stem}.semantic.glb"
if semantic_glb_candidate.exists():
semantic_glb_path = str(semantic_glb_candidate)
# 方法2: 在语义标注目录中查找
# 从 glb-v0.2 推断到 semantic-annots-v0.2
if 'glb-v0.2' in str(mesh_dir):
semantic_dir = str(mesh_dir).replace('glb-v0.2', 'semantic-annots-v0.2')
semantic_dir_obj = Path(semantic_dir)
if semantic_dir_obj.exists():
semantic_txt_candidate = semantic_dir_obj / f"{mesh_stem}.semantic.txt"
if semantic_txt_candidate.exists():
semantic_txt_path = str(semantic_txt_candidate)
semantic_glb_candidate = semantic_dir_obj / f"{mesh_stem}.semantic.glb"
if semantic_glb_candidate.exists():
semantic_glb_path = str(semantic_glb_candidate)
# 方法3: 在上级目录的语义标注目录中查找
if semantic_txt_path is None and semantic_glb_path is None:
# 尝试在dataset_HM3D目录下查找
dataset_root = mesh_dir.parent.parent if 'glb-v0.2' in str(mesh_dir) else mesh_dir.parent
semantic_annots_dir = dataset_root / "hm3d-example-semantic-annots-v0.2"
if semantic_annots_dir.exists():
# 查找场景目录
for scene_dir in semantic_annots_dir.iterdir():
if scene_dir.is_dir():
semantic_txt_candidate = scene_dir / f"{mesh_stem}.semantic.txt"
if semantic_txt_candidate.exists():
semantic_txt_path = str(semantic_txt_candidate)
break
semantic_glb_candidate = scene_dir / f"{mesh_stem}.semantic.glb"
if semantic_glb_candidate.exists():
semantic_glb_path = str(semantic_glb_candidate)
break
# 优先返回semantic.txt
if semantic_txt_path:
return semantic_txt_path
elif semantic_glb_path:
return semantic_glb_path
else:
return None
def load_semantic_info(mesh_path: str) -> Tuple[Optional[Dict[int, List[int]]], Optional[str]]:
"""
加载语义信息（优先使用semantic.txt，如果不存在则使用semantic.glb）
Args:
mesh_path: GLB文件路径
Returns:
(room_to_objects, semantic_file_path):
- room_to_objects: {room_id: [object_ids]} 字典，如果不存在则返回None
- semantic_file_path: 使用的语义文件路径，如果不存在则返回None
"""
semantic_file_path = find_semantic_file(mesh_path)
if semantic_file_path is None:
return None, None
# 根据文件类型选择解析方法
if semantic_file_path.endswith('.txt'):
room_to_objects = parse_semantic_txt(semantic_file_path)
return room_to_objects, semantic_file_path
elif semantic_file_path.endswith('.glb'):
room_to_objects = load_semantic_from_glb(semantic_file_path)
return room_to_objects, semantic_file_path
else:
return None, None
|