dmlc/gluon-cv

Any easier tutorial for custom object detection?

Open

#281 opened on Aug 28, 2018

View on GitHub
 (3 comments) (0 reactions) (0 assignees)Python (5,610 stars) (1,224 forks)batch import
good first issue

Description

First of all, many thanks for the great tutorials for object detection using pre-trained models. The code is so simple and easier to follow.

The problem is that custom object detection (with a small dataset) and transfer learning tutorials are complicated for beginners. I saw a tutorial somewhere that shows custom object detection using just 5 lines of code. Can you write a simple tutorial for us?

This looks very complicated. https://gluon-cv.mxnet.io/build/examples_classification/transfer_learning_minc.html Do we really need to learn all these details while you have such a great high level API?

I am sorry if this is not the right place to post this question but I need your help on this.

Thanks again.

Contributor guide

Tech stack
python
Domain
documentationmachine learning
Issue type
documentation
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
2
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
half day
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
stale
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
clear
Prerequisites
basic knowledge of object detectionfamiliarity with Python
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
80
Research direction
The issue requests a simpler tutorial for custom object detection using transfer learning. The existing transfer learning example at https://gluon cv.mxnet.io/build/examples classification/transfer learning minc.html is considered too complicated. The goal is to create a concise tutorial using the high level API, possibly in 5 lines of code. Research the current GluonCV API for object detection, identify the simplest way to load a pretrained model and fine tune on a small custom dataset, and write a step by step guide. No linked PRs or maintainer responses are present.