[Support]: "Object Lifecycle" screenshot resolution stretched to portrait.

[ryank3nn3dy]

Describe the problem you are having

some of the snapshot images in the object lifecycle are in a stretched forced portrait orientation. I have done the below in troubleshooting.

1. Originally only had a single input, which was the main stream. Resolution was matching the feed
2. Created a secondary input for each camera, split the detect to the sub-stream and matched the resolution, no change
3. Checked the main stream camera feed via VLC RTSP, is landscape
4. Checked the sub-stream camera feed via VLC RTSP, is landscape

I've attached 3 supporting screenshots

1. F12 developer tools of the image showing the INCORRECT image resolution
2. screenshot of object lifecycle showing correct image size and orientation
3. screenshot of object lifecycle showing INCORRECT image size and orientation (no change in config taken 1 hr apart)
4. Camera config showing sub-stream

I know my config file is a bit of a mess, still been tinkering and doing intitial setup.

Version

0.16.0-c2f8de9

What browser(s) are you using?

Chrome

Frigate config file

mqtt:
  enabled: false

version: 0.16-0

tls:
  enabled: false

#GOOGLE API
#genai:
  #enabled: true
  #provider: gemini
  #api_key: 
  #model: gemini-1.5-flash

genai:
  enabled: true
  provider: azure_openai
  base_url: 
    https://URL
  api_key: 
    KEY
  prompt: Analyze the {label} in these images from the {camera} security camera.
    Focus on the actions, behavior, and potential intent of the {label}, rather 
    than just describing its appearance.
  object_prompts:
    person: Examine the main person in these images. Give a brief description of
      them, What are they doing and what might their actions suggest about their
      intent (e.g., approaching a door, leaving an area, standing still)? Do not
      describe the surroundings or static details.
    car: Observe the primary vehicle in these images. Focus on its movement, 
      direction, or purpose (e.g., parking, approaching, circling). If it's a 
      delivery vehicle, mention the company.
    dog: Observe the dog or dogs in these images. Is it on a lead with a 
      handler? Give a brief description of them, What are they doing and what 
      might their actions suggest about their intent. Do not describe the 
      surroundings or static details.
    motorcycle: Observe the primary motorcycle in these images. Focus on its 
      movement, direction, or purpose (e.g., parking, approaching, circling) 
      What kind of motorcyle is it?

ffmpeg:
  hwaccel_args: preset-vaapi

detectors:
  ov_0:
    type: openvino
    device: GPU
  ov_1:
    type: openvino
    device: GPU
model:
  width: 300
  height: 300
  input_tensor: nhwc
  input_pixel_format: bgr
  path: /openvino-model/ssdlite_mobilenet_v2.xml
  labelmap_path: /openvino-model/coco_91cl_bkgr.txt

record:
  enabled: true
  retain:
    days: 7
    mode: motion
  alerts:
    retain:
      days: 30
  detections:
    retain:
      days: 30

snapshots:
  enabled: true
  retain:
    default: 30

review:
  # Optional: alerts configuration
  #alerts:
    # Optional: enables alerts for the camera (default: shown below)
    #enabled: true
     #Optional: labels that qualify as an alert (default: shown below)
    #labels:
    #  - car
    #  - person
    #  - bicycle
    #  - motorcycle
    #  - cat
    #  - dog
    #  - shoe
    # Optional: required zones for an object to be marked as an alert (default: none)
    # NOTE: when settings required zones globally, this zone must exist on all cameras
    #       or the config will be considered invalid. In that case the required_zones
    #       should be configured at the camera level.
    #required_zones:
    #  - driveway
  # Optional: detections configuration
  detections:
    # Optional: enables detections for the camera (default: shown below)
    enabled: true
    # Optional: labels that qualify as a detection (default: all labels that are tracked / listened to)
    labels:
      - car
      - person
      - bicycle
      - motorcycle
      - cat
      - dog
      - shoe
      - potted plant
objects:
  track:
    - person
    - car
    - bicycle
    - motorcycle
    - cat
    - dog
  filters:
    person:
      threshold: 0.79
    motorcycle:
      threshold: 0.70
    dog:
      threshold: 0.70
detect:
  enabled: true
  stationary:
    # Optional: Frequency for confirming stationary objects (default: same as threshold)
    # When set to 1, object detection will run to confirm the object still exists on every frame.
    # If set to 10, object detection will run to confirm the object still exists on every 10th frame.
    interval: 50
    # Optional: Number of frames without a position change for an object to be considered stationary (default: 10x the frame rate or 10s)
    threshold: 50
    # Optional: Define a maximum number of frames for tracking a stationary object (default: not set, track forever)
    # This can help with false positives for objects that should only be stationary for a limited amount of time.
    # It can also be used to disable stationary object tracking. For example, you may want to set a value for person, but leave
    # car at the default.
    # WARNING: Setting these values overrides default behavior and disables stationary object tracking.
    #          There are very few situations where you would want it disabled. It is NOT recommended to
    #          copy these values from the example config into your config unless you know they are needed.
    max_frames:
      # Optional: Default for all object types (default: not set, track forever)
      default: 3000000
      # Optional: Object specific values
      objects:
        person: 1000
        car: 1000000
cameras:
  Driveway:
    detect:
      enabled: true # <---- turn on detection
      width: 960
      height: 576
      fps: 10
    ffmpeg:
      inputs:
        - path: rtsp://user:PW@192.168.1.2:554/Streaming/Channels/102
          roles:
            - detect
        - path: rtsp://user:PW@192.168.1.2:554/Streaming/Channels/101
          roles:
            - record
    motion:
      mask:
        - 1,0.226,0.896,0.164,0.798,0.115,0.716,0.094,0.618,0.056,0.498,0.022,0.341,0.028,0.223,0.044,0.14,0.062,0,0.113,0,0,0.21,0,0.394,0,0.577,0,0.73,0,0.859,0,1,0
        - 0.055,0.348,0.111,0.289,0.182,0.253,0.238,0.241,0.283,0.277,0.305,0.328,0.315,0.389,0.319,0.505,0.311,0.589,0.282,0.625,0.227,0.655,0.196,0.688,0.167,0.678
      threshold: 50
      contour_area: 50
      improve_contrast: true
    zones:
      Visitor_Parking:
        coordinates: 0.65,0.26,0.994,0.397,0.998,0.835,0.856,0.996,0.661,1
        loitering_time: 0
      Driveways:
        coordinates: 
          0.781,0.136,0.647,0.068,0.335,0.103,0.11,0.143,0.037,0.22,0.064,0.36,0.124,0.293,0.238,0.255,0.286,0.288,0.31,0.385,0.314,0.498,0.313,0.592,0.279,0.687,0.218,0.742,0.558,0.98,0.569,0.652,0.577,0.366,0.621,0.214,0.68,0.172
        loitering_time: 0
        inertia: 3
    objects:
      filters:
        car:
          mask: 0,0.951,0.623,0.951,0.624,1,0,1
    notifications:
      enabled: false
  Frontdoor:
    detect:
      enabled: true # <---- turn on detection
      width: 960
      height: 576
      fps: 10
    ffmpeg:
      inputs:
        - path: rtsp://user:PW@192.168.1.2:554/Streaming/Channels/202
          roles:
            - detect
        - path: rtsp://user:PW@192.168.1.2:554/Streaming/Channels/201
          roles:
            - record
    motion:
      mask:
        - 0.081,0.388,0.171,0.539,0.292,0.38,0.245,0.229,0.171,0.232,0.113,0.287,0.076,0.334
        - 0.004,0.267,0.005,0.22,0.013,0.192,0.026,0.182,0.051,0.182,0.061,0.197,0.06,0.215
        - 0.506,0.192,0.574,0.157,0.589,0.085,0.597,0.008,0.542,0.003,0.483,0.004
    zones:
      FrontDoorArea:
        coordinates: 
          0.172,0.571,0.371,0.273,0.476,0.357,0.569,0.24,0.591,0.106,0.602,0.016,0.742,0.002,0.837,0.002,0.872,0.011,0.915,0.233,0.941,0.357,0.675,0.954,0.658,0.991,0.015,0.991,0.004,0.405
        loitering_time: 0
      MediaRoomEntry:
        coordinates: 
          0.174,0.559,0.004,0.28,0.139,0.13,0.195,0.011,0.321,0.007,0.38,0.249
        loitering_time: 0
    notifications:
      enabled: true
  Sidegate:
    detect:
      enabled: true # <---- turn on detection
      width: 960
      height: 576
      fps: 10
    ffmpeg:
      inputs:
        - path: rtsp://user:PW@192.168.1.2:554/Streaming/Channels/302
          roles:
            - detect
        - path: rtsp://user:PW@192.168.1.2:554/Streaming/Channels/301
          roles:
            - record
    motion:
      mask: 
        0.334,0.159,0.453,0.18,0.453,0.111,0.476,0.08,0.474,0.004,0.054,0.005,0.004,0.071,0.002,0.919,0.086,0.555,0.165,0.401,0.245,0.278
    zones:
      Carport:
        coordinates: 
          0.405,0.385,0.462,0.474,0.688,0.672,0.78,0.513,0.843,0.57,0.879,0.498,0.553,0.209,0.443,0.335
        loitering_time: 0
      Sidepath:
        coordinates: 
          0.482,0.266,0.325,0.183,0.177,0.407,0.109,0.549,0.046,0.863,0.026,0.989,0.602,0.996,0.679,0.853,0.642,0.756,0.676,0.687,0.392,0.385
        loitering_time: 0
    notifications:
      enabled: false
    objects:
      filters:
        car:
          mask: 0.783,0.502,0.621,0.764,0.721,0.873,0.892,0.587
  Garages:
    detect:
      enabled: true # <---- turn on detection
      width: 960
      height: 576
      fps: 10
    ffmpeg:
      inputs:
        - path: rtsp://user:PW@192.168.1.2:554/Streaming/Channels/602
          roles:
            - detect
        - path: rtsp://user:PW@192.168.1.2:554/Streaming/Channels/601
          roles:
            - record
    motion:
      mask:
        - 0.676,0.046,0.685,0.168,0.753,0.183,0.753,0.072
        - 0,0.415,0.058,0.349,0.094,0.309,0.139,0.26,0.182,0.219,0.246,0.169,0.291,0.131,0.347,0.095,0.401,0.066,0.468,0.035,0.509,0.018,0.55,0,0.11,0.003,0.001,0.004
    zones:
      Carport:
        coordinates: 
          0.894,0.584,0.596,0.462,0.597,0.427,0.199,0.335,0.166,0.399,0.146,0.485,0.121,0.577,0.106,0.65,0.092,0.777,0.257,1,0.995,0.991,0.984,0.172,0.913,0.588
        loitering_time: 0
      MainDriveway:
        coordinates: 
          0.08,0.332,0.122,0.275,0.162,0.24,0.581,0.099,0.942,0.222,0.899,0.573,0.607,0.46,0.607,0.418,0.206,0.322,0.176,0.339,0.131,0.344
        loitering_time: 0

      FrontDoor:
        coordinates: 0.935,0.209,0.933,0.017,0.807,0.008,0.801,0.166
        loitering_time: 0
    notifications:
      enabled: false
semantic_search:
  enabled: true
  model_size: large
face_recognition:
  enabled: true
  model_size: large
lpr:
  enabled: true
classification:
  bird:
    enabled: false
camera_groups:
  HilltopBirdseye:
    order: 1
    icon: LuAtom
    cameras: birdseye
notifications:
  enabled: false
  email:

Relevant Frigate log output

N/A

Relevant go2rtc log output

N/A

FFprobe output from your camera

Stream data is obtained with ffprobe.

Stream 1
Video:
Codec: H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10
Resolution: 960x480 (2/1 aspect ratio)
FPS: 10/1
Stream 2
Video:
Codec: H.265 / HEVC (High Efficiency Video Coding)
Resolution: 1280x720 (16/9 aspect ratio)
FPS: 40/1

Frigate stats

No response

Install method

Docker Compose

docker-compose file or Docker CLI command

docker run
  -d
  --name='frigate'
  --net='bridge'
  --privileged=true
  -e TZ="Australia/Brisbane"
  -e HOST_OS="Unraid"
  -e HOST_HOSTNAME="CUBE"
  -e HOST_CONTAINERNAME="frigate"
  -e 'FRIGATE_RTSP_PASSWORD'='PW'
  -e 'PLUS_API_KEY'=''
  -e 'LIBVA_DRIVER_NAME'='iHD'
  -e 'NVIDIA_VISIBLE_DEVICES'=''
  -e 'NVIDIA_DRIVER_CAPABILITIES'=''
  -e 'YOLO_MODELS'=''
  -e 'USE_FP16'=''
  -e 'TRT_MODEL_PREP_DEVICE'=''
  -l net.unraid.docker.managed=dockerman
  -l net.unraid.docker.webui='http://[IP]:[PORT:8971]'
  -l net.unraid.docker.icon='https://raw.githubusercontent.com/yayitazale/unraid-templates/main/frigate.png'
  -p '8971:8971/tcp'
  -p '8554:8554/tcp'
  -p '5000:5000/tcp'
  -p '8555:8555/tcp'
  -p '8555:8555/udp'
  -v '/mnt/user/appdata/frigate/':'/config':'rw'
  -v '/mnt/user/data/media/Frigate/':'/media/frigate':'rw'
  -v '/etc/localtime':'/etc/localtime':'rw'
  --device='/dev/bus/usb'
  --device='/dev/dri/renderD128'
  --shm-size=256m
  --mount type=tmpfs,target=/tmp/cache,tmpfs-size=1000000000
  --restart unless-stopped 'ghcr.io/blakeblackshear/frigate:stable'

Object Detector

OpenVino

Network connection

Wired

Camera make and model

annke/hikvision

Screenshots of the Frigate UI's System metrics pages

No response

Any other information that may be helpful

No response

[inkeep-ai-assistant[bot]]

The object lifecycle images in Frigate are generated from the frames of the camera stream defined by the detect role during the object's tracking period (1). The resolution and quality of these images depend on your camera's detect stream configuration.

Looking at your configuration, you have different resolutions between your main stream (1280x720) and sub stream (960x480) (1). The object lifecycle images are captured from whichever stream is assigned the detect role - in your case, the sub stream at 960x480 resolution.

The stretched portrait orientation you're seeing could be related to aspect ratio handling when the detect stream resolution doesn't match the camera's native aspect ratio. Your FFprobe output shows the sub stream has a 960x480 resolution with a 2:1 aspect ratio, while the main stream is 1280x720 with a 16:9 aspect ratio.

To improve the quality and correct orientation of object lifecycle images, consider:

1. Adjust detect stream resolution: The documentation recommends that cameras should be configured directly for the desired resolutions and frame rates you want to use in Frigate (1). Configure your camera's sub stream to maintain the proper aspect ratio.

2. Match detect resolution in config: Ensure your detect resolution settings match your camera's actual sub stream output (1):

detect:
  width: 960
  height: 480  # Match your actual sub stream resolution

The ideal resolution for detection is one where the objects you want to detect fit inside the dimensions of the model used by Frigate (320x320), and larger resolutions do improve performance if the objects are very small in the frame (1).

📚 Sources:

1. Camera setup

---

Have another question? Just tag @inkeep.

[ryank3nn3dy]

ok, thank you for the information!

I've disabled H264+ on all cameras (NVR wanted a reboot after each camera change).

Will keep an eye on this and see how it goes.

Thank you!

[ryank3nn3dy]

I just had some events, and the images were stretched still..... :(

[NickM-27]

I'm having a hard time seeing where the issue is in the original screenshots, they look like two different cameras?

[ryank3nn3dy]

shit, apologies... it was the wrong screenshot. i've edited and included the correct one.

[hawkeye217]

I'm not sure what can be done here as your issue isn't consistently reproduced. All Frigate does to grab the screenshot is run an ffmpeg command on the copied record stream:

frigate/frigate/util/image.py

Line 941 in 15729e0

def get_image_from_recording(

My best guess is that something is inconsistent enough with your camera stream that ffmpeg can't determine the correct aspect ratio when it resizes it. We've not had any other reports of this, either.

You might consider running the streams through go2rtc (with restreaming) to see if somehow that can "clean" them up. The upside is that you will get a vastly improved live view.

@inkeep can probably refactor your entire config to use go2rtc with restreaming very quickly.

[inkeep-ai-assistant[bot]]

I don't have specific documentation in the available sources that explicitly states which stream (detect or record) is used for Object Lifecycle images. The sources I have access to don't contain detailed information about how Object Lifecycle images are generated or which stream they're sourced from.

To properly troubleshoot this issue, you would need to check your record stream's aspect ratio. Looking at your configuration, your record role is assigned to:

• rtsp://user:PW@192.168.1.2:554/Streaming/Channels/101 for the Driveway camera

According to your FFprobe output, you have two streams with different aspect ratios:

• Stream 1: 960x480 (2:1 aspect ratio)
• Stream 2: 1280x720 (16:9 aspect ratio)

If the Object Lifecycle images are indeed sourced from the record stream as you mentioned, the aspect ratio mismatch between your streams could be causing the stretching issue.

I'd recommend checking the documentation at GitHub for more specific information about Object Lifecycle image generation, as this detail isn't covered in the sources I have available.

---

Have another question? Just tag @inkeep.

[inkeep-ai-assistant[bot]]

Based on your configuration and the issue you're experiencing, I can help you set up go2rtc with restreaming to potentially resolve the aspect ratio issues with your Object Lifecycle images. This approach reduces connections to your cameras and can help normalize the streams(1).

Here's how to refactor your configuration to use go2rtc restreaming:

go2rtc:
  streams:
    driveway:
      - rtsp://user:PW@192.168.1.2:554/Streaming/Channels/101  # main stream
      - "ffmpeg:driveway#video=h264#hardware"  # transcode if needed
    driveway_sub:
      - rtsp://user:PW@192.168.1.2:554/Streaming/Channels/102  # sub stream
    
    frontdoor:
      - rtsp://user:PW@192.168.1.2:554/Streaming/Channels/201  # main stream
      - "ffmpeg:frontdoor#video=h264#hardware"  # transcode if needed
    frontdoor_sub:
      - rtsp://user:PW@192.168.1.2:554/Streaming/Channels/202  # sub stream
    
    sidegate:
      - rtsp://user:PW@192.168.1.2:554/Streaming/Channels/301  # main stream
      - "ffmpeg:sidegate#video=h264#hardware"  # transcode if needed
    sidegate_sub:
      - rtsp://user:PW@192.168.1.2:554/Streaming/Channels/302  # sub stream
    
    garages:
      - rtsp://user:PW@192.168.1.2:554/Streaming/Channels/601  # main stream
      - "ffmpeg:garages#video=h264#hardware"  # transcode if needed
    garages_sub:
      - rtsp://user:PW@192.168.1.2:554/Streaming/Channels/602  # sub stream

cameras:
  Driveway:
    ffmpeg:
      inputs:
        - path: rtsp://127.0.0.1:8554/driveway_sub
          input_args: preset-rtsp-restream
          roles:
            - detect
        - path: rtsp://127.0.0.1:8554/driveway
          input_args: preset-rtsp-restream
          roles:
            - record
    # ... rest of your camera config
  
  Frontdoor:
    ffmpeg:
      inputs:
        - path: rtsp://127.0.0.1:8554/frontdoor_sub
          input_args: preset-rtsp-restream
          roles:
            - detect
        - path: rtsp://127.0.0.1:8554/frontdoor
          input_args: preset-rtsp-restream
          roles:
            - record
    # ... rest of your camera config

(1)

This configuration creates a single connection to each camera stream, then restreams it locally for both detect and record roles(1). The restreaming can help normalize any inconsistencies in your camera streams that might be causing the aspect ratio issues.

Since you're using H.265 streams, I've included the ffmpeg transcoding option with hardware acceleration to convert to H.264 if needed(2). This should help with compatibility and potentially resolve the aspect ratio problems you're experiencing.

📚 Sources:

1. Restream
2. Configuring go2rtc

---

Have another question? Just tag @inkeep.

[hawkeye217]

@inkeep be sure to include the live --> streams directives in your example so that go2rtc is used for live view.

[inkeep-ai-assistant[bot]]

You're absolutely right! Here's the complete configuration including the live streams directives to ensure go2rtc is used for live view and optimal performance(1):

go2rtc:
  streams:
    driveway:
      - rtsp://user:PW@192.168.1.2:554/Streaming/Channels/101  # main stream
      - "ffmpeg:driveway#video=h264#hardware"  # transcode if needed
    driveway_sub:
      - rtsp://user:PW@192.168.1.2:554/Streaming/Channels/102  # sub stream
    
    frontdoor:
      - rtsp://user:PW@192.168.1.2:554/Streaming/Channels/201  # main stream
      - "ffmpeg:frontdoor#video=h264#hardware"  # transcode if needed
    frontdoor_sub:
      - rtsp://user:PW@192.168.1.2:554/Streaming/Channels/202  # sub stream
    
    sidegate:
      - rtsp://user:PW@192.168.1.2:554/Streaming/Channels/301  # main stream
      - "ffmpeg:sidegate#video=h264#hardware"  # transcode if needed
    sidegate_sub:
      - rtsp://user:PW@192.168.1.2:554/Streaming/Channels/302  # sub stream
    
    garages:
      - rtsp://user:PW@192.168.1.2:554/Streaming/Channels/601  # main stream
      - "ffmpeg:garages#video=h264#hardware"  # transcode if needed
    garages_sub:
      - rtsp://user:PW@192.168.1.2:554/Streaming/Channels/602  # sub stream

cameras:
  Driveway:
    ffmpeg:
      inputs:
        - path: rtsp://127.0.0.1:8554/driveway_sub
          input_args: preset-rtsp-restream
          roles:
            - detect
        - path: rtsp://127.0.0.1:8554/driveway
          input_args: preset-rtsp-restream
          roles:
            - record
    live:
      streams:
        Main Stream: driveway
        Sub Stream: driveway_sub
    # ... rest of your camera config
  
  Frontdoor:
    ffmpeg:
      inputs:
        - path: rtsp://127.0.0.1:8554/frontdoor_sub
          input_args: preset-rtsp-restream
          roles:
            - detect
        - path: rtsp://127.0.0.1:8554/frontdoor
          input_args: preset-rtsp-restream
          roles:
            - record
    live:
      streams:
        Main Stream: frontdoor
        Sub Stream: frontdoor_sub
    # ... rest of your camera config
  
  Sidegate:
    ffmpeg:
      inputs:
        - path: rtsp://127.0.0.1:8554/sidegate_sub
          input_args: preset-rtsp-restream
          roles:
            - detect
        - path: rtsp://127.0.0.1:8554/sidegate
          input_args: preset-rtsp-restream
          roles:
            - record
    live:
      streams:
        Main Stream: sidegate
        Sub Stream: sidegate_sub
    # ... rest of your camera config
  
  Garages:
    ffmpeg:
      inputs:
        - path: rtsp://127.0.0.1:8554/garages_sub
          input_args: preset-rtsp-restream
          roles:
            - detect
        - path: rtsp://127.0.0.1:8554/garages
          input_args: preset-rtsp-restream
          roles:
            - record
    live:
      streams:
        Main Stream: garages
        Sub Stream: garages_sub
    # ... rest of your camera config

(1)(2)

The live streams configuration allows manual selection of streams in the Live UI and provides dropdown options for choosing between different streams per device(1). This configuration also ensures you get the highest quality live view experience using go2rtc's MSE or WebRTC streaming instead of the limited jsmpeg stream(1).

Since you're using H.265 streams and have already disabled H.265+, the go2rtc restreaming with hardware transcoding should help normalize any stream inconsistencies that might be causing the aspect ratio issues with your Object Lifecycle images(3).

📚 Sources:

1. Live View
2. Restream
3. Configuring go2rtc

---

Have another question? Just tag @inkeep.

[ryank3nn3dy]

@inkeep these are now failing.... from the Go2Rtc logs;

error="streams: parse "rtsp://admin:PW@192.168.1.2:554/Streaming/Channels/201": invalid URL escape "%ej", exec/rtsp\n[rtsp @ 0x55f4562f6400] method DESCRIBE failed: 404 Not Found\n[in#0 @ 0x55f4562f60c0] Error opening input: Server returned 404 Not Found\nError opening input file rtsp://127.0.0.1:8554/frontdoor?video&source=ffmpeg:frontdoor%23video%3Dh264%23hardware.\nError opening input files: Server returned 404 Not Found\n" stream=frontdoor

From the Frigate logs;
error | 2025-09-10 13:59:30 | ffmpeg.Garages.detect | Error opening input file rtsp://192.168.1.134:8554/garages_sub.
error | 2025-09-10 13:59:30 | ffmpeg.Garages.detect | Error opening input files: Server returned 404 Not Found
info | 2025-09-10 13:59:30 | watchdog.Garages | Restarting ffmpeg...

I don't understand why it is trying to connect to rtsp://127.0.0.1:8554 when i have changed the IP twice now in troubleshooting.

It was throwing errors, so i though the loopback might not work (docker container in Unraid) so i used the assigned internal IP, same errror, then i tried the bridged unraid network IP, didn't work either....

[ryank3nn3dy]

found the issue... i have a % in my password....

From Chat

Working now. Will keep an eye on the detection and orientations now.

[ryank3nn3dy]

@hawkeye217 restreams are all setup and working properly, but i just had a detection and the image was squished still.....

i don't get it....

[ryank3nn3dy]

To all who helped and anyone here in the future.....

I ended up watching the RTSP feed in VLC and noticed that each time the cameras internal motion detection was triggered, the feed would switch to a squashed resolution.

The annke cameras have the ability to trigger the recording of other cameras based off motion on any other cameras, this is setup in "Events -> Linkage Method -> Trigger Channel". I had other cameras set, and for whatever reason this was causing the odd issue in the RTSP feed.

i turned off all the triggers, saved each camera and the issue is now fixed, no more odd resolutions in the feeds.

thank you all for the support.

[NickM-27]

Thanks for updating with the answer