From c09aee85e60dd21b12505afd1e0115461c4b692d Mon Sep 17 00:00:00 2001 From: Nate Meyer Date: Wed, 31 Aug 2022 01:23:59 -0400 Subject: [PATCH] Add documentation for new model config parameters --- docs/docs/configuration/advanced.md | 14 ++++++++++++++ docs/docs/configuration/index.md | 5 +++++ 2 files changed, 19 insertions(+) diff --git a/docs/docs/configuration/advanced.md b/docs/docs/configuration/advanced.md index b6b2ba150..c2e93138a 100644 --- a/docs/docs/configuration/advanced.md +++ b/docs/docs/configuration/advanced.md @@ -50,6 +50,20 @@ database: If using a custom model, the width and height will need to be specified. +Custom models may also require different input tensor formats. The colorspace conversion supports RGB, BGR, or YUV frames to be sent to the object detector. The input tensor shape parameter is a list of axis to let the detector reorder (transpose) the standard layout [BHWC] to match what specified by the model. + +| Tensor Dimension | Description | +| :--------------: | -------------- | +| B | Batch Size | +| H | Model Height | +| W | Model Width | +| C | Color Channels | + +```yaml +model: + input_tensor: ["B", "H", "W", "C"] +``` + The labelmap can be customized to your needs. A common reason to do this is to combine multiple object types that are easily confused when you don't need to be as granular such as car/truck. By default, truck is renamed to car because they are often confused. You cannot add new object types, but you can change the names of existing objects in the model. ```yaml diff --git a/docs/docs/configuration/index.md b/docs/docs/configuration/index.md index 57bc815dd..e4dd7bbfd 100644 --- a/docs/docs/configuration/index.md +++ b/docs/docs/configuration/index.md @@ -97,6 +97,11 @@ model: width: 320 # Required: Object detection model input height (default: shown below) height: 320 + # Optional: Object detection model input colorspace + # Valid values are rgb, bgr, or yuv. (default: shown below) + input_pixel_format: rgb + # Optional: Object detection model input tensor format (default: shown below) + input_tensor: ["B", "H", "W", "C"] # Optional: Label name modifications. These are merged into the standard labelmap. labelmap: 2: vehicle