open-mmlab/mmaction2

[PR Welcome] More data augmentation methods

Open

#436 opened on Dec 12, 2020

View on GitHub
 (10 comments) (5 reactions) (1 assignee)Python (3,648 stars) (1,137 forks)batch import
discussionhelp wantedopen research

Description

Describe the feature More data augmentation methods.

Motivation When training our own models, we need to TRY EVERYTHING...

Additional context

  • MMaction2 existing data augmentation methods:
    • random flip.
    • all kinds of crop and resize
    • color jitter(brightness/contrast/saturation/hue)

Plans

More image augmentation methods could refer to PaddleClas.

Contributor guide

Tech stack
pythonpytorch
Domain
datamachine learning
Issue type
feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-2 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
clear
Prerequisites
PythonPyTorchvideo data augmentation
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
60
Research direction
Start by examining the existing data augmentation pipeline in mmaction2, particularly the transforms module. Review how previous augmentations like imgaug and mixup were integrated (PR #492, #681). Then implement support for albumentations following the imgaug pattern. For videomix, study the paper and existing mixup implementation. Refer to PaddleClas documentation for additional ideas.