Learning Data Augmentation Strategies for Object Detection — mentioned by rocauc

Fri, 03 Apr 2020 18:41:01 GMT

As others have noted, this is data augmentation, and it's incredibly useful to increase variation in training data to help decrease overfitting. It's not a silver bullet. It won't capture the natural variations that happen in the real world. But new forms of augmentation (like OP) are helping us get closer. For example, MixMatch creates "mosaic" images by combining images across the training set [1]. In object detection, bounding box only augmentations are improving models by introducing variation [2]. And an anecdote: I work on https://roboflow.ai , and we've seen customers make…

Paper and discussion history. A mention is not an endorsement.

Learning Data Augmentation Strategies for Object Detection — HN submission

Thu, 27 Jun 2019 08:54:29 GMT

Learning Data Augmentation Strategies for Object Detection

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Learning Data Augmentation Strategies for Object Detection · HackerPapers discussions

Learning Data Augmentation Strategies for Object Detection — mentioned by rocauc

Learning Data Augmentation Strategies for Object Detection — HN submission