Deep Representation Learning for Domain Adaptation of Semantic Image Segmentation

05/10/2018
by   Assia Benbihi, et al.
0

Deep Convolutional Neural Networks have pushed the state-of-the art for semantic segmentation provided that a large amount of images together with pixel-wise annotations is available. Data collection is expensive and a solution to alleviate it is to use transfer learning. This reduces the amount of annotated data required for the network training but it does not get rid of this heavy processing step. We propose a method of transfer learning without annotations on the target task for datasets with redundant content and distinct pixel distributions. Our method takes advantage of the approximate content alignment of the images between two datasets when the approximation error prevents the reuse of annotation from one dataset to another. Given the annotations for only one dataset, we train a first network in a supervised manner. This network autonomously learns to generate deep data representations relevant to the semantic segmentation. Then the images in the new dataset, we train a new network to generate a deep data representation that matches the one from the first network on the previous dataset. The training consists in a regression between feature maps and does not require any annotations on the new dataset. We show that this method reaches performances similar to a classic transfer learning on the PASCAL VOC dataset with synthetic transformations.

READ FULL TEXT

page 2

page 6

page 7

page 8

research
09/04/2017

Dataset Augmentation with Synthetic Images Improves Semantic Segmentation

Although Deep Convolutional Neural Networks trained with strong pixel-le...
research
06/19/2020

BEV-Seg: Bird's Eye View Semantic Segmentation Using Geometry and Semantic Point Cloud

Bird's-eye-view (BEV) is a powerful and widely adopted representation fo...
research
09/23/2020

Interactive Learning for Semantic Segmentation in Earth Observation

Dense pixel-wise classification maps output by deep neural networks are ...
research
03/21/2023

Full or Weak annotations? An adaptive strategy for budget-constrained annotation campaigns

Annotating new datasets for machine learning tasks is tedious, time-cons...
research
12/04/2018

PolyMapper: Extracting City Maps using Polygons

We propose a method to leapfrog pixel-wise, semantic segmentation of (ae...
research
12/24/2022

HandsOff: Labeled Dataset Generation With No Additional Human Annotations

Recent work leverages the expressive power of generative adversarial net...
research
10/14/2020

Data compression to choose a proper dynamic network representation

Dynamic network data are now available in a wide range of contexts and d...

Please sign up or login with your details

Forgot password? Click here to reset