TRTPlus

---

Introduction

• Easy to Use
• Fast Inference
• Automatic Model Decoding

---

Support

Only Testing On Linux.

Task	Work
segmentation	✔️
yolov8 classification	✔️
yolov8 detection	✔️
yolov8 segmentation	❔
yolov8 obb detection	✔️
yolov10 detection	✔️
yolov11 detection	✔️

Install

cd <your work dir>
git clone https://github.com/akira4O4/tensorrt-lab.git
cd tensorrt-lab

---

How to use

Step1: Write your label.txt

If your project task is asegmentation,the label.txt first line must be 0_background_.

E.g. classification_label.txt

classes1
classes2
...
classes7
classes8

E.g. segmentation_label.txt

0_background_
1_xxx
2_xxx
...
5_xxx
6_xxx

Step2: Modify Your main.cpp

Refer to the cpp file in the examples folder to write your code

examples/
├── yolov10_det.cpp
├── yolov8_cls.cpp
├── yolov8_det.cpp
└── yolov8_seg.cpp

---

Compile & Rund

Attention: Configure your CmakeLists.txt

mkdir build && cd build
cmake ..
sudo make -j8

./main.bin