MCPcopy Create free account

hub / github.com/AI-FanGe/OpenAIglasses_for_Navigation / functions

Functions403 in github.com/AI-FanGe/OpenAIglasses_for_Navigation

⨍ Functions 403◇ Types & classes 31↳ Endpoints 3
↓ 1 callersMethoddrawPolygon
(points, color, lineWidth = 2)
static/vision_renderer.js:348
↓ 1 callersMethoddrawProgressBars
(tracking, width, height)
static/visualizer.js:375
↓ 1 callersMethoddrawSegmentMode
(width, height)
static/visualizer.js:158
↓ 1 callersMethoddrawStats
(width, height)
static/visualizer.js:438
↓ 1 callersMethoddrawStatusMessage
(message)
static/vision_renderer.js:424
↓ 1 callersMethoddrawTargetMarker
(x, y)
static/vision_renderer.js:352
↓ 1 callersMethoddrawTargetMarker
(x, y)
static/visualizer.js:475
↓ 1 callersMethoddrawTargetObject
(obj)
static/vision.js:265
↓ 1 callersMethoddrawTechProgressBars
(alignScore, rangeScore)
static/vision_renderer.js:268
↓ 1 callersMethoddrawTrackMode
(width, height)
static/visualizer.js:257
↓ 1 callersFunctiondraw_command_pill
统一改为右上角白色文案。不再绘制底部圆角按钮。
yolomedia.py:80
↓ 1 callersFunctiondraw_hand_contour
绘制手部landmarks的凸包轮廓
yolomedia.py:489
↓ 1 callersFunctiondraw_hands_mono
(img_bgr, hand_lms, color=(0, 255, 255), r=2, t=2)
yolomedia.py:187
↓ 1 callersFunctionestimate_translation_flow
估计两帧之间的平移光流 :param prev_gray: 前一帧灰度图 :param curr_gray: 当前帧灰度图 :param mask: 可选的掩码 :return: (中位光流幅度, 平移矩阵)
utils.py:124
↓ 1 callersFunctionextract_english_label
返回 (label_en, source)；source ∈ {'local', 'qwen', 'fallback'}
qwen_extractor.py:33
↓ 1 callersFunctionformatTime
(timestamp = Date.now())
static/main.js:154
↓ 1 callersFunctiongen
()
audio_stream.py:134
↓ 1 callersMethodgetModeText
(mode)
static/vision.js:195
↓ 1 callersFunctionget_center_guidance
判断物体是否在画面中心，返回引导方向 返回: (方向文字, 是否已居中)
yolomedia.py:612
↓ 1 callersMethodget_compression_stats
获取压缩统计信息
audio_compressor.py:418
↓ 1 callersFunctionget_guidance_direction
根据手心和物体中心位置，以及面积比，返回引导方向 返回: (方向文字, 是否需要前后调整)
yolomedia.py:543
↓ 1 callersMethodget_visualization_data
获取可视化数据 返回包含所有可视化元素的字典
crosswalk_awareness.py:295
↓ 1 callersFunctiongpu_infer_slot
统一管理 GPU 并发限流 + inference_mode + AMP autocast
obstacle_detector_client.py:37
↓ 1 callersMethodhistory
(self)
navigation_master.py:53
↓ 1 callersFunctionimu_broadcast
(msg: str)
app_main.py:1089
↓ 1 callersMethodinitUI
()
static/vision.js:24
↓ 1 callersFunctioninit_model
初始化YOLO模型（单帧处理模式）
trafficlight_detection.py:477
↓ 1 callersMethodinit_traffic_light_detector
初始化红绿灯检测器
workflow_blindpath.py:263
↓ 1 callersFunctionkeepalive_loop
()
app_main.py:759
↓ 1 callersMethodload_and_compress
加载并压缩音频文件（统一转换为8kHz）
audio_compressor.py:322
↓ 1 callersFunctionload_navigation_models
加载盲道导航所需的模型
app_main.py:119
↓ 1 callersFunctionload_wav_file
加载WAV文件并返回PCM数据（自动转换为8kHz）
audio_player.py:65
↓ 1 callersFunctionmain
红绿灯检测主函数 参数: headless: 是否无头模式（不显示OpenCV窗口） stop_event: threading.Event，用于停止检测
trafficlight_detection.py:184
↓ 1 callersFunctionmain
(headless: bool = False, prompt_name: str = None, stop_event=None)
yolomedia.py:637
↓ 1 callersMethodmajority
(self)
navigation_master.py:43
↓ 1 callersFunctionnorm_name
(s: str)
yolomedia.py:203
↓ 1 callersMethodon_voice_command
(self, text: str)
navigation_master.py:353
↓ 1 callersMethodpcm16_to_adpcm
将16位PCM转换为4位ADPCM 压缩率：75%（16bit -> 4bit） 保持较好的语音质量
audio_compressor.py:105
↓ 1 callersMethodpcm16_to_ulaw
将16位PCM转换为8位μ-law 压缩率：50%（16bit -> 8bit）
audio_compressor.py:20
↓ 1 callersFunctionpost
(coro)
app_main.py:797
↓ 1 callersFunctionpreload_all_audio
预加载所有音频文件到内存
audio_player.py:136
↓ 1 callersMethodprocess_frame
(self, bgr: np.ndarray)
navigation_master.py:420
↓ 1 callersFunctionprocess_imu_and_maybe_store
(d: Dict[str, Any])
app_main.py:1130
↓ 1 callersFunctionprocess_queue
()
audio_player.py:183
↓ 1 callersFunctionregister_stream_route
(app)
audio_stream.py:120
↓ 1 callersMethodrenderFlashAnimation
(progress)
static/vision_renderer.js:250
↓ 1 callersMethodrenderFrame
(data)
static/vision_renderer.js:99
↓ 1 callersMethodreset
重置状态
workflow_crossstreet.py:277
↓ 1 callersMethodreset
重置导航器状态
workflow_blindpath.py:3115
↓ 1 callersFunctionset_current_command
(text: str)
yolomedia.py:72
↓ 1 callersMethodset_text_classes
(self, names: List[str])
yoloe_backend.py:23
↓ 1 callersMethodsetupCanvas
()
static/vision_renderer.js:36
↓ 1 callersMethodsetupCanvas
()
static/visualizer.js:36
↓ 1 callersFunctionshouldShowTimestamp
()
static/main.js:145
↓ 1 callersFunctionsplit_combined_voice
将组合语音拆分为多个独立语音 例如："远处发现斑马线,在画面左侧" → ["远处发现斑马线", "在画面左侧"]
crosswalk_awareness.py:333
↓ 1 callersMethodstartRenderLoop
()
static/vision_renderer.js:81
↓ 1 callersFunctionstart_ai_with_text
硬重置后，开启新的 AI 语音输出。
app_main.py:611
↓ 1 callersFunctionstart_ai_with_text_custom
扩展版的AI启动函数，支持识别特殊命令
app_main.py:410
↓ 1 callersMethodstart_blind_path_navigation
启动盲道导航模式
navigation_master.py:293
↓ 1 callersMethodstart_crossing
启动过马路模式
navigation_master.py:307
↓ 1 callersMethodstart_item_search
启动找物品模式，暂停当前导航
navigation_master.py:323
↓ 1 callersMethodstart_traffic_light_detection
启动红绿灯检测模式
navigation_master.py:314
↓ 1 callersFunctionstart_yolomedia_with_target
启动yolomedia线程，搜索指定物品
app_main.py:358
↓ 1 callersFunctionstop_current_recognition
()
asr_core.py:79
↓ 1 callersMethodstop_item_search
停止找物品模式
navigation_master.py:335
↓ 1 callersFunctionstream_chat
发起一轮 Omni-Turbo ChatCompletions 流式对话： - content_list: OpenAI chat 的 content，多模态（image_url/text） - 以 stream=True 返回 - 增量产出：OmniStreamP
omni_client.py:27
↓ 1 callersFunctionui_reset_overlay
每帧调用一次，重置叠加行计数
trafficlight_detection.py:91
↓ 1 callersFunctionui_reset_overlay
每帧调用一次，重置叠加行计数（改为右上角布局）。
yolomedia.py:55
↓ 1 callersMethodulaw_to_pcm16
将8位μ-law转换回16位PCM
audio_compressor.py:37
↓ 1 callersFunctionupdateDataPanel
(roll, pitch, yaw, gx, gy, gz, ax, ay, az)
static/main.js:438
↓ 1 callersMethodupdateDataPanel
(data)
static/vision.js:188
↓ 1 callersFunctionupdateLighting
()
static/main.js:617
↓ 1 callersMethodupdateProgress
(data)
static/vision.js:174
↓ 1 callersMethodupdateStatus
(data)
static/vision.js:154
↓ 1 callersFunctionwarp_mask
使用仿射变换对掩码进行变换 :param mask: 输入掩码 :param M: 2x3的仿射变换矩阵 :param output_shape: 输出形状 (width, height) :return: 变换后的掩码
utils.py:101
MethodICM42688
compile/ICM42688.cpp:4
Method__init__
初始化录制器 :param output_dir: 输出目录 :param fps: 视频帧率（默认15fps）
sync_recorder.py:18
Method__init__
(self, model_path: Optional[str] = None, device: Optional[Union[str, int]] = None)
yoloe_backend.py:18
Method__init__
(self, seg_model=None, coco_model=None, obs_model=None, device_id: str = "esp32")
workflow_crossstreet.py:205
Method__init__
(self, yolo_model)
workflow_crossstreet.py:1794
Method__init__
(self)
workflow_crossstreet.py:1810
Method__init__
( self, on_sdk_error: Callable[[str], None], post: Callable[[asyncio.Future], None],
asr_core.py:99
Method__init__
compression_type: "none", "ulaw", "adpcm"
audio_compressor.py:313
Method__init__
(self, text_delta: Optional[str] = None, audio_b64: Optional[str] = None)
omni_client.py:23
Method__init__
(self)
crosswalk_awareness.py:19
Method__init__
(self, size: int = 8)
navigation_master.py:37
Method__init__
(self)
navigation_master.py:67
Method__init__
(self, blind_nav: BlindPathNavigator, cross_nav: CrossStreetNavigator,
navigation_master.py:246
Method__init__
(self, enable=True)
yolomedia.py:208
Method__init__
(self, model_path: str = 'models/yoloe-11l-seg.pt')
obstacle_detector_client.py:50
Method__init__
初始化导航器 :param yolo_model: YOLO分割模型（可选） :param obstacle_detector: 障碍物检测器（可选）
workflow_blindpath.py:89
Method__post_init__
(self)
workflow_crossstreet.py:117
Method__post_init__
(self)
workflow_blindpath.py:81
Function_audio_worker
音频播放工作线程
audio_player.py:162
Method_binarize
(mask)
workflow_blindpath.py:911
Function_draw_badge
(img, text, pos=(10, 28), fg="white", bg="blue")
navigation_master.py:219
Method_draw_data_panel
绘制数据面板（需要Pillow）
workflow_blindpath.py:3065
Function_draw_frame_border
(img, color=(0,255,0), thickness=3)
navigation_master.py:233
Function_draw_progress_bar
(img, ratio: float, pos=(10, 90), size=(180, 10), color="cyan")
navigation_master.py:237
Function_draw_state_panel
(img, kv: Dict[str, Any], pos=(10, 60))
navigation_master.py:227
← previousnext →201–300 of 403, ranked by callers