frigate: Presence Detection not working relibly

Describe the problem you are having

I am trying to utilise the new presence detection to determine if my little one is in his cot. Frigate consistently identifies him with a score of between 75 and 80%. The bounding box appears to consistently report him, however in the home assistant device the detected object seems to vary between 0 and 1 inconsistently, with no apparent reason.

Im new to Frigate, so may be missing something!

Version

10.0.0

Frigate config file

{
  "birdseye": {
    "enabled": true,
    "height": 720,
    "mode": "objects",
    "quality": 8,
    "width": 1280
  },
  "cameras": {
    "nursery": {
      "best_image_timeout": 60,
      "detect": {
        "enabled": true,
        "fps": 1,
        "height": 270,
        "max_disappeared": 5,
        "stationary": {
          "interval": 0,
          "max_frames": {
            "default": null,
            "objects": {}
          },
          "threshold": 10
        },
        "width": 480
      },
      "ffmpeg": {
        "global_args": [
          "-hide_banner",
          "-loglevel",
          "warning"
        ],
        "hwaccel_args": [],
        "input_args": [
          "-avoid_negative_ts",
          "make_zero",
          "-fflags",
          "+genpts+discardcorrupt",
          "-rtsp_transport",
          "tcp",
          "-stimeout",
          "5000000",
          "-use_wallclock_as_timestamps",
          "1"
        ],
        "inputs": [
          {
            "global_args": [],
            "hwaccel_args": [],
            "input_args": [],
            "path": "rtsp://192.168.1.1:7447/xK9LDPNt3sUe7Jnu",
            "roles": [
              "record",
              "rtmp",
              "detect"
            ]
          }
        ],
        "output_args": {
          "detect": [
            "-f",
            "rawvideo",
            "-pix_fmt",
            "yuv420p"
          ],
          "record": [
            "-f",
            "segment",
            "-segment_time",
            "10",
            "-segment_format",
            "mp4",
            "-reset_timestamps",
            "1",
            "-strftime",
            "1",
            "-c",
            "copy",
            "-an"
          ],
          "rtmp": [
            "-c",
            "copy",
            "-f",
            "flv"
          ]
        }
      },
      "ffmpeg_cmds": [
        {
          "cmd": "ffmpeg -hide_banner -loglevel warning -avoid_negative_ts make_zero -fflags +genpts+discardcorrupt -rtsp_transport tcp -stimeout 5000000 -use_wallclock_as_timestamps 1 -i rtsp://192.168.1.1:7447/xK9LDPNt3sUe7Jnu -c copy -f flv rtmp://127.0.0.1/live/nursery -r 1 -s 480x270 -f rawvideo -pix_fmt yuv420p pipe:",
          "roles": [
            "record",
            "rtmp",
            "detect"
          ]
        }
      ],
      "live": {
        "height": 720,
        "quality": 8
      },
      "motion": {
        "contour_area": 30,
        "delta_alpha": 0.2,
        "frame_alpha": 0.2,
        "frame_height": 50,
        "mask": "",
        "threshold": 25
      },
      "mqtt": {
        "bounding_box": true,
        "crop": true,
        "enabled": true,
        "height": 270,
        "quality": 70,
        "required_zones": [],
        "timestamp": true
      },
      "name": "nursery",
      "objects": {
        "filters": {
          "person": {
            "mask": null,
            "max_area": 24000000,
            "min_area": 0,
            "min_score": 0.5,
            "threshold": 0.7
          }
        },
        "mask": "",
        "track": [
          "person"
        ]
      },
      "record": {
        "enabled": false,
        "events": {
          "max_seconds": 300,
          "objects": null,
          "post_capture": 5,
          "pre_capture": 5,
          "required_zones": [],
          "retain": {
            "default": 10,
            "mode": "motion",
            "objects": {}
          }
        },
        "expire_interval": 60,
        "retain": {
          "days": 0,
          "mode": "all"
        },
        "retain_days": null
      },
      "rtmp": {
        "enabled": true
      },
      "snapshots": {
        "bounding_box": true,
        "clean_copy": true,
        "crop": false,
        "enabled": false,
        "height": null,
        "quality": 70,
        "required_zones": [],
        "retain": {
          "default": 10,
          "mode": "motion",
          "objects": {}
        },
        "timestamp": false
      },
      "timestamp_style": {
        "color": {
          "blue": 255,
          "green": 255,
          "red": 255
        },
        "effect": null,
        "format": "%m/%d/%Y %H:%M:%S",
        "position": "tl",
        "thickness": 2
      },
      "zones": {
        "cot": {
          "coordinates": [
            "180,1",
            "310,1",
            "360,270",
            "135,270"
          ],
          "filters": {
            "person": {
              "mask": null,
              "max_area": 24000000,
              "min_area": 0,
              "min_score": 0.4,
              "threshold": 0.5
            }
          },
          "objects": [
            "person"
          ]
        }
      }
    }
  },
  "database": {
    "path": "/media/frigate/frigate.db"
  },
  "detect": {
    "enabled": true,
    "fps": 5,
    "height": 720,
    "max_disappeared": null,
    "stationary": {
      "interval": 0,
      "max_frames": {
        "default": null,
        "objects": {}
      },
      "threshold": null
    },
    "width": 1280
  },
  "detectors": {
    "cpu": {
      "device": "usb",
      "num_threads": 3,
      "type": "cpu"
    }
  },
  "environment_vars": {},
  "ffmpeg": {
    "global_args": [
      "-hide_banner",
      "-loglevel",
      "warning"
    ],
    "hwaccel_args": [],
    "input_args": [
      "-avoid_negative_ts",
      "make_zero",
      "-fflags",
      "+genpts+discardcorrupt",
      "-rtsp_transport",
      "tcp",
      "-stimeout",
      "5000000",
      "-use_wallclock_as_timestamps",
      "1"
    ],
    "output_args": {
      "detect": [
        "-f",
        "rawvideo",
        "-pix_fmt",
        "yuv420p"
      ],
      "record": [
        "-f",
        "segment",
        "-segment_time",
        "10",
        "-segment_format",
        "mp4",
        "-reset_timestamps",
        "1",
        "-strftime",
        "1",
        "-c",
        "copy",
        "-an"
      ],
      "rtmp": [
        "-c",
        "copy",
        "-f",
        "flv"
      ]
    }
  },
  "live": {
    "height": 720,
    "quality": 8
  },
  "logger": {
    "default": "info",
    "logs": {}
  },
  "model": {
    "height": 320,
    "labelmap": {},
    "labelmap_path": null,
    "path": null,
    "width": 320
  },
  "motion": null,
  "mqtt": {
    "client_id": "frigate",
    "host": "192.168.1.14",
    "password": "mqtt567",
    "port": 1883,
    "stats_interval": 60,
    "tls_ca_certs": null,
    "tls_client_cert": null,
    "tls_client_key": null,
    "tls_insecure": null,
    "topic_prefix": "frigate",
    "user": "mqtt"
  },
  "objects": {
    "filters": null,
    "mask": "",
    "track": [
      "person"
    ]
  },
  "record": {
    "enabled": false,
    "events": {
      "max_seconds": 300,
      "objects": null,
      "post_capture": 5,
      "pre_capture": 5,
      "required_zones": [],
      "retain": {
        "default": 10,
        "mode": "motion",
        "objects": {}
      }
    },
    "expire_interval": 60,
    "retain": {
      "days": 0,
      "mode": "all"
    },
    "retain_days": null
  },
  "rtmp": {
    "enabled": true
  },
  "snapshots": {
    "bounding_box": true,
    "clean_copy": true,
    "crop": false,
    "enabled": false,
    "height": null,
    "quality": 70,
    "required_zones": [],
    "retain": {
      "default": 10,
      "mode": "motion",
      "objects": {}
    },
    "timestamp": false
  },
  "timestamp_style": {
    "color": {
      "blue": 255,
      "green": 255,
      "red": 255
    },
    "effect": null,
    "format": "%m/%d/%Y %H:%M:%S",
    "position": "tl",
    "thickness": 2
  }
}

Relevant log output

None

FFprobe output from your camera

N/A

Frigate stats

No response

Operating system

HassOS

Install method

HassOS Addon

Coral version

CPU (no coral)

Network connection

Wireless

Camera make and model

Unifi G3 Instant

Any other information that may be helpful

No response

About this issue

  • Original URL
  • State: open
  • Created 2 years ago
  • Comments: 29 (1 by maintainers)

Most upvoted comments

Try reducing the time in the history area to see how quickly those drops are going off for. I’d be guessing it is something along the lines of what @NickM-27 has previously mentioned where it is somewhat of an instant off then back on because Frigate has momentarily stopped detecting the motion and sent it through to HA, but not enough for the event to stop within Frigate itself.

I’m having a similar issue where once a person is detected, binary_sensor_person_motion_detection stays on and debug shows no boundary box for a detected person once they’ve left the FOV. I’m not using zones or anything besides dog, cat and person motion detection. This is only happening on one of my Reolink E1 Zoom cameras.

Just curious, if you check the logs do you have any connection issues with the camera or is everything working as expected? I’m assuming the event ended at the appropriate time?

I am actually having connection problems with one of them, but not the one in question. The one with connection issues works fine. The events start and end at the appropriate times but it seems like if I spend an extended period of time in the garage the sensor hangs even though Frigate reports no people.

I’m having a similar issue where once a person is detected, binary_sensor_person_motion_detection stays on and debug shows no boundary box for a detected person once they’ve left the FOV. I’m not using zones or anything besides dog, cat and person motion detection. This is only happening on one of my Reolink E1 Zoom cameras.

Okay thanks, last question: Do you have frigate events enabled for this zone? If so do you see one long zone or does it get split into events like this graph from HA?

I don’t, no, I was purely using for presence detection.