Megvii-BaseDetection / Megvii-BaseDetection/YOLOX

YOLOX on NVIDIA DeepStream

Open
#1,599 0 comments 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
10.6k
Forks
2.5k
PR merge metrics
No merged PRs in 30d

Description

Repository to use YOLOv8 model on NVIDIA DeepStream SDK without ONNX conversion.

Link: https://github.com/marcoslucianops/DeepStream-Yolo

Config
board = NVIDIA Tesla V100 (AWS: p3.2xlarge)
batch-size = 1
eval = val2017 (COCO)
sample = 1920x1080 video
NMS config
  • Eval
nms-iou-threshold=0.65
pre-cluster-threshold=0.001
topk=300
  • Test
nms-iou-threshold=0.45
pre-cluster-threshold=0.25
topk=300
Results (comming soon)
DeepStream Precision Resolution IoU=0.5:0.95 IoU=0.5 IoU=0.75 FPS
(without display)
YOLOX-x standard FP16 640
YOLOX-l standard FP16 640
YOLOX-m standard FP16 640
YOLOX-s standard FP16 640
YOLOX-Tiny standard FP16 416
YOLOX-Nano standard FP16 416
YOLOX-x legacy FP16 640
YOLOX-l legacy FP16 640
YOLOX-m legacy FP16 640
YOLOX-s legacy FP16 640
YOLOX-Tiny legacy FP16 416
YOLOX-Nano legacy FP16 416

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reading the issue's linked DeepStream-Yolo repository and compare its YOLOX integration with this repository's model and export documentation. The issue does not name YOLOX files or tests; completion would need a defined integration target and the requested DeepStream evaluation results filled into the benchmark table.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
ai, computer-vision
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.