Skip to content

tensorrt_yolox#

Purpose#

This package detects target objects e.g., cars, trucks, bicycles, and pedestrians on a image based on YOLOX model.

Inner-workings / Algorithms#

Cite#

Zheng Ge, Songtao Liu, Feng Wang, Zeming Li, Jian Sun, "YOLOX: Exceeding YOLO Series in 2021", arXiv preprint arXiv:2107.08430, 2021 [ref]

Inputs / Outputs#

Input#

Name Type Description
in/image sensor_msgs/Image The input image

Output#

Name Type Description
out/objects tier4_perception_msgs/DetectedObjectsWithFeature The detected objects with 2D bounding boxes
out/image sensor_msgs/Image The image with 2D bounding boxes for visualization

Parameters#

Core Parameters#

Name Type Default Value Description
score_threshold float 0.3 If the objectness score is less than this value, the object is ignored in yolox layer.
nms_threshold float 0.7 The IoU threshold for NMS method

NOTE: These two parameters are only valid for "plain" model (described later).

Node Parameters#

Name Type Default Value Description
model_path string "" The onnx file name for yolox model
label_path string "" The label file with label names for detected objects written on it
precision string "fp16" The inference mode: "fp32", "fp16", "int8"
build_only bool false shutdown node after TensorRT engine file is built
calibration_algorithm string "MinMax" Calibration algorithm to be used for quantization when precision==int8. Valid value is one of: Entropy",("Legacy" | "Percentile"), "MinMax"]
dla_core_id int -1 If positive ID value is specified, the node assign inference task to the DLA core
quantize_first_layer bool false If true, set the operating precision for the first (input) layer to be fp16. This option is valid only when precision==int8
quantize_last_layer bool false If true, set the operating precision for the last (output) layer to be fp16. This option is valid only when precision==int8
profile_per_layer bool false If true, profiler function will be enabled. Since the profile function may affect execution speed, it is recommended to set this flag true only for development purpose.
clip_value double 0.0 If positive value is specified, the value of each layer output will be clipped between [0.0, clip_value]. This option is valid only when precision==int8 and used to manually specify the dynamic range instead of using any calibration
preprocess_on_gpu bool true If true, pre-processing is performed on GPU
calibration_image_list_path string "" Path to a file which contains path to images. Those images will be used for int8 quantization.

Assumptions / Known limits#

The label contained in detected 2D bounding boxes (i.e., out/objects) will be either one of the followings:

  • CAR
  • PEDESTRIAN ("PERSON" will also be categorized as "PEDESTRIAN")
  • BUS
  • TRUCK
  • BICYCLE
  • MOTORCYCLE

If other labels (case insensitive) are contained in the file specified via the label_file parameter, those are labeled as UNKNOWN, while detected rectangles are drawn in the visualization result (out/image).

Onnx model#

A sample model (named yolox-tiny.onnx) is downloaded automatically during the build process. To accelerate Non-maximum-suppression (NMS), which is one of the common post-process after object detection inference, EfficientNMS_TRT module is attached after the ordinal YOLOX (tiny) network. The EfficientNMS_TRT module contains fixed values for score_threshold and nms_threshold in it, hence these parameters are ignored when users specify ONNX models including this module.

This package accepts both EfficientNMS_TRT attached ONNXs and models published from the official YOLOX repository (we referred to them as "plain" models).

In addition to yolox-tiny.onnx, a custom model named yolox-sPlus-opt.onnx is either available. This model is based on YOLOX-s and tuned to perform more accurate detection with almost comparable execution speed with yolox-tiny. To get better results with this model, users are recommended to use some specific running arguments such as precision:=int8, calibration_algorithm:=Entropy, clip_value:=6.0. Users can refer launch/yolox_sPlus_opt.launch.xml to see how this model can be used.

All models are automatically converted to TensorRT format. These converted files will be saved in the same directory as specified ONNX files with .engine filename extension and reused from the next run. The conversion process may take a while (typically 10 to 20 minutes) and the inference process is blocked until complete the conversion, so it will take some time until detection results are published (even until appearing in the topic list) on the first run

Package acceptable model generation#

To convert users' own model that saved in PyTorch's pth format into ONNX, users can exploit the converter offered by the official repository. For the convenience, only procedures are described below. Please refer the official document for more detail.

For plain models#

  1. Install dependency

    git clone git@github.com:Megvii-BaseDetection/YOLOX.git
cd YOLOX
python3 setup.py develop --user

  2. Convert pth into ONNX

    python3 tools/export_onnx.py \
  --output-name YOUR_YOLOX.onnx \
  -f YOUR_YOLOX.py \
  -c YOUR_YOLOX.pth

For EfficientNMS_TRT embedded models#

  1. Install dependency

    git clone git@github.com:Megvii-BaseDetection/YOLOX.git
cd YOLOX
python3 setup.py develop --user
pip3 install git+ssh://git@github.com/wep21/yolox_onnx_modifier.git --user

  2. Convert pth into ONNX

    python3 tools/export_onnx.py \
  --output-name YOUR_YOLOX.onnx \
  -f YOUR_YOLOX.py \
  -c YOUR_YOLOX.pth
  --decode_in_inference

  3. Embed EfficientNMS_TRT to the end of YOLOX

    yolox_onnx_modifier YOUR_YOLOX.onnx -o YOUR_YOLOX_WITH_NMS.onnx

Label file#

A sample label file (named label.txt)is also downloaded automatically during the build process (NOTE: This file is incompatible with models that output labels for the COCO dataset (e.g., models from the official YOLOX repository)).

This file represents the correspondence between class index (integer outputted from YOLOX network) and class label (strings making understanding easier). This package maps class IDs (incremented from 0) with labels according to the order in this file.

Reference repositories#