Sitemap
Nerd For Tech

NFT is an Educational Media House. Our mission is to bring the invaluable knowledge and experiences of experts from all over the world to the novice. To know more about us, visit https://www.nerdfortech.org/.

Follow publication

Review — Rethinking ImageNet Pre-training (Object Detection, Semantic Segmentation)

4 min readFeb 21, 2021

--

The model, ResNet50-FPN Using GN, trained from random initialization needs more iterations to converge, but converges to a solution that is no worse than the fine-tuning counterpart.
@ Medium)

Outline

1. Number of Training Images & Setup

1.1. Number of Training Images Involved

Total numbers of images, instances, and pixels seen during all training iterations, for pre-training + fine-tuning (green bars) vs. from random initialization (purple bars).

1.2. Setup

2. Training from Scratch to Match Accuracy

Learning curves of APbbox on COCO val2017 using Mask R-CNN with R101-FPN and GN

3. Training from Scratch with Less Data

Training with 10k COCO images

4. Discussions

4.1. Is ImageNet pre-training necessary?

4.2. Is ImageNet Useful?

4.3. Is Big Data Helpful?

4.4. Shall We Pursuit Universal Representations?

--

--

Nerd For Tech
Nerd For Tech

Published in Nerd For Tech

NFT is an Educational Media House. Our mission is to bring the invaluable knowledge and experiences of experts from all over the world to the novice. To know more about us, visit https://www.nerdfortech.org/.

Sik-Ho Tsang
Sik-Ho Tsang

Written by Sik-Ho Tsang

PhD, Researcher. I share what I learn. :) Linktree: https://linktr.ee/shtsang for Twitter, LinkedIn, etc.

No responses yet