Effective Train/Test Stratification for Object Detection

Fri, 30 Sep 2022 19:47:39 +0530

TL;DR: Here’s a talk based on this post:

There’s an unavoidable, inherent difficulty in fine-tuning deep neural networks, which stems from the lack of training data. It would seem ridiculous to a layperson that a pretrained vision model (containing millions of parameters, trained on millions of images) could learn to solve highly specific problems. Therefore, when fine-tuned models do perform well, they seem all the more miraculous. But on the other hand, we also know that it is easier to move from the general to the specific, than the reverse. Specialization after generalization is easier than the reverse.

Weighted Loss Functions for Instance Segmentation

Fri, 17 Aug 2018 14:52:19 +0530

This post is a follow up to my talk, Practical Image Classification & Object Detection at PyData Delhi 2018. You can watch the talk here:

and see the slides here.

I spoke at length about the different kinds of problems in computer vision and how they are interpreted in deep learning architectures. I spent a fair bit of time on instance and semantic segmentation (for an introduction to these problems, watch Justin Johnson’s lecture from the Stanford CS231 course here). In short, semantic segmentation deals with designating each pixel of an image as belonging to a class, and instance segmentation deals with identifying which instance of a class each pixel belongs to. The following images show an example of semantic versus instance segmentation.

Computervision on Jaidev's Blog

Effective Train/Test Stratification for Object Detection

Weighted Loss Functions for Instance Segmentation