albumentations
BlenderProc
Our great sponsors
albumentations | BlenderProc | |
---|---|---|
28 | 15 | |
13,395 | 2,544 | |
1.9% | 3.5% | |
8.3 | 8.3 | |
7 days ago | 11 days ago | |
Python | Python | |
MIT License | GNU General Public License v3.0 only |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
albumentations
-
Augment specific classes?
You can use albumentations if you are comfortable with using open source libraries https://github.com/albumentations-team/albumentations
-
Ask HN: What side projects landed you a job?
One of the members of the core team of our open-source library https://albumentations.ai/
It was not the only reason he was hired; it was a solid addition to his already good performance at the interviews.
Or at least that is what the hiring manager later said.
-
The Lack of Compensation in Open Source Software Is Unsustainable
I am one of the creators and maintainers of https://albumentations.ai/.
- 12800+ stars
-
Burn Deep Learning Framework Release 0.7.0: Revamped (de)serialization, optimizer & module overhaul, initial ONNX support and tons of new features.
Is something planned to support data augmentations? Something like https://albumentations.ai/
-
How to label augmented images for training YOLO algorithm?
Here you go: https://albumentations.ai/
-
Unstable Diffusion bounces back with $19,000 raised in one day, by using Stripe
I think they should use some data augmentation techniques like I am using for Infinity AI if you wanna see more here. Note that most of these do not work for image generation.
-
Tokyo Drift : detecting drift in images with NannyML and Whylogs
Our second approach was a more automated one. Here the idea was to try out an image augmentation library, Albumentations, and use it for adversarial attacks. This time, instead of one-shot images, we applied the transformations at random time ranges. We chose for these transformations also to be more subtle than then one-shot images, such as vertical flips, grayscaling, downscaling, …
-
[D] Improve machine learning with same number of images
Check out albumentations. If your use case is segmentation, check out the offline augmentation of this project
-
What are the best programs/scripts for image augmentation of YOLO5 training dataset. Something like roboflow but free)
I think this is the most popular open source project: https://github.com/albumentations-team/albumentations
-
To get dataset for face image restoration.
You can also curate your own dataset by using open source images (https://universe.roboflow.com/search?q=faces%20images%3E1000) and open source augmentations (https://github.com/albumentations-team/albumentations). Or you can do use the augmentation UI (https://docs.roboflow.com/image-transformations/image-augmentation) to apply noise, blurring, shear, crop, etc.
BlenderProc
-
Synthetic image Generation
Blender with add-ons (Kubric, BlenderProc)
-
dataset collection for transfer learning
If you are interested, there are open source solutions on top of Unity, Blender, Unreal. You can generate yourself the data you described easier than it looks (the amount of options and settings can be intimidating with these tools).
-
Are there any tools to generate images and labels from 3d models/games?
Blender addons like https://github.com/google-research/kubric and https://github.com/DLR-RM/BlenderProc
- How to get started with synthetic data generation?
-
Course/Learning Material recommendations for getting started with Synthetic Data Generation for Computer Vision Models
I have been going through some papers and reviewing existing methods and I've come across stuff like UnrealCV (https://unrealcv.org/) and blenderproc (https://github.com/DLR-RM/BlenderProc).
-
Searching for MIT CSAIL's IKEA dataset
I'm trying to use BlenderProc to automatically generate training data for object recognition.
- [P] BlenderProc2: Photorealistic Rendering of Procedurally Generated Scenes
-
[D] What's the best method to generate synthetic data for an image with text? Small dataset
Check this out https://github.com/DLR-RM/BlenderProc. I haven't used it extensively, but it seems to decent for generating synthetic image data.
- Apple’s Machine Learning Team Introduces ‘Hypersim’: A Photorealistic Synthetic Dataset for Holistic Indoor Scene Understanding
-
Using Blender for Computer Vision
That looks really interesting, have you seen BlenderProc: https://github.com/DLR-RM/BlenderProc it looks really similar just that BlenderProc already supports a vast variety of datasets and is fully documented.
What are some alternatives?
imgaug - Image augmentation for machine learning experiments.
zpy - Synthetic data for computer vision. An open source toolkit using Blender and Python.
YOLO-Mosaic - Perform mosaic image augmentation on data for training a YOLO model
com.unity.perception - Perception toolkit for sim2real training and validation in Unity
labelme2coco - A lightweight package for converting your labelme annotations into COCO object detection format.
segmentation_models.pytorch - Segmentation models with pretrained backbones. PyTorch.
autoalbument - AutoML for image augmentation. AutoAlbument uses the Faster AutoAugment algorithm to find optimal augmentation policies. Documentation - https://albumentations.ai/docs/autoalbument/
agi2nerf - Simple tool for converting Agisoft XML files to NERF JSON files for https://github.com/NVlabs/instant-ngp
Mask-RCNN-TF2 - Mask R-CNN for object detection and instance segmentation on Keras and TensorFlow 2.0
segmentation_models - Segmentation models with pretrained backbones. Keras and TensorFlow Keras.
ttach - Image Test Time Augmentation with PyTorch!
SingleViewReconstruction - Official Code: 3D Scene Reconstruction from a Single Viewport