From 89b54f19c80e4dc97fa4d769c178a3933f97191b Mon Sep 17 00:00:00 2001
From: Nicolas Mowen <nickmowen213@gmail.com>
Date: Fri, 18 Apr 2025 16:51:04 -0600
Subject: [PATCH] Add YOLOv9 support to RKNN (#17791)

* Add yolov9

* Undo

* Update docs for rknn yolov9

* Update docs notes

* Add infernece times table
---
 docs/docs/configuration/object_detectors.md | 23 +++++++++++++++++++++
 docs/docs/frigate/hardware.md               |  6 ++++++
 frigate/detectors/plugins/rknn.py           |  3 +++
 3 files changed, 32 insertions(+)

diff --git a/docs/docs/configuration/object_detectors.md b/docs/docs/configuration/object_detectors.md
index 8ba4b463f..fa6132dc8 100644
--- a/docs/docs/configuration/object_detectors.md
+++ b/docs/docs/configuration/object_detectors.md
@@ -849,6 +849,7 @@ The inference time was determined on a rk3588 with 3 NPU cores.
 | deci-fp16-yolonas_s | 24         | 25                   |
 | deci-fp16-yolonas_m | 62         | 35                   |
 | deci-fp16-yolonas_l | 81         | 45                   |
+| yolov9_tiny         | 8          | 35                   |
 | yolox_nano          | 3          | 16                   |
 | yolox_tiny          | 6          | 20                   |
 
@@ -864,7 +865,9 @@ model: # required
   # - deci-fp16-yolonas_s
   # - deci-fp16-yolonas_m
   # - deci-fp16-yolonas_l
+  # your yolonas_model.rknn
   path: deci-fp16-yolonas_s
+  model_type: yolonas
   width: 320
   height: 320
   input_pixel_format: bgr
@@ -878,6 +881,24 @@ The pre-trained YOLO-NAS weights from DeciAI are subject to their license and ca
 
 :::
 
+#### YOLO (v9)
+
+```yaml
+model: # required
+  # name of model (will be automatically downloaded) or path to your own .rknn model file
+  # possible values are:
+  # - yolov9-t
+  # - yolov9-s
+  # your yolo_model.rknn
+  path: /config/model_cache/rknn_cache/yolov9-t.rknn
+  model_type: yolo-generic
+  width: 320
+  height: 320
+  input_tensor: nhwc
+  input_dtype: float
+  labelmap_path: /labelmap/coco-80.txt
+```
+
 #### YOLOx
 
 ```yaml
@@ -886,7 +907,9 @@ model: # required
   # possible values are:
   # - yolox_nano
   # - yolox_tiny
+  # your yolox_model.rknn
   path: yolox_tiny
+  model_type: yolox
   width: 416
   height: 416
   input_tensor: nhwc
diff --git a/docs/docs/frigate/hardware.md b/docs/docs/frigate/hardware.md
index 15e3b1e0a..061e1f770 100644
--- a/docs/docs/frigate/hardware.md
+++ b/docs/docs/frigate/hardware.md
@@ -165,6 +165,12 @@ Frigate supports hardware video processing on all Rockchip boards. However, hard
 - RK3576
 - RK3588
 
+| Name            | YOLOv9 Inference Time | YOLO-NAS Inference Time     | YOLOx Inference Time      |
+| --------------- | --------------------- | --------------------------- | ------------------------- |
+| rk3588 3 cores  | ~ 35 ms               | small: ~ 20 ms med: ~ 30 ms | nano: 18 ms tiny: 20 ms   |
+| rk3566 1 core   |                       | small: ~ 96 ms              |                           |
+
+
 The inference time of a rk3588 with all 3 cores enabled is typically 25-30 ms for yolo-nas s.
 
 ## What does Frigate use the CPU for and what does it use a detector for? (ELI5 Version)
diff --git a/frigate/detectors/plugins/rknn.py b/frigate/detectors/plugins/rknn.py
index b71de2ad6..5b573cd56 100644
--- a/frigate/detectors/plugins/rknn.py
+++ b/frigate/detectors/plugins/rknn.py
@@ -11,6 +11,7 @@ from pydantic import Field
 from frigate.const import MODEL_CACHE_DIR
 from frigate.detectors.detection_api import DetectionApi
 from frigate.detectors.detector_config import BaseDetectorConfig, ModelTypeEnum
+from frigate.util.model import post_process_yolo
 
 logger = logging.getLogger(__name__)
 
@@ -284,6 +285,8 @@ class Rknn(DetectionApi):
     def post_process(self, output):
         if self.detector_config.model.model_type == ModelTypeEnum.yolonas:
             return self.post_process_yolonas(output)
+        elif self.detector_config.model.model_type == ModelTypeEnum.yologeneric:
+            return post_process_yolo(output, self.width, self.height)
         elif self.detector_config.model.model_type == ModelTypeEnum.yolox:
             return self.post_process_yolox(output, self.grids, self.expanded_strides)
         else: