Do Different Deep Metric Learning Losses Lead to Similar Learned Features?

05/05/2022
by   Konstantin Kobs, et al.
106

Recent studies have shown that many deep metric learning loss functions perform very similarly under the same experimental conditions. One potential reason for this unexpected result is that all losses let the network focus on similar image regions or properties. In this paper, we investigate this by conducting a two-step analysis to extract and compare the learned visual features of the same model architecture trained with different loss functions: First, we compare the learned features on the pixel level by correlating saliency maps of the same input images. Second, we compare the clustering of embeddings for several image properties, e.g. object color or illumination. To provide independent control over these properties, photo-realistic 3D car renders similar to images in the Cars196 dataset are generated. In our analysis, we compare 14 pretrained models from a recent study and find that, even though all models perform similarly, different loss functions can guide the model to learn different features. We especially find differences between classification and ranking based losses. Our analysis also shows that some seemingly irrelevant properties can have significant influence on the resulting embedding. We encourage researchers from the deep metric learning community to use our methods to get insights into the features learned by their proposed methods.

READ FULL TEXT

page 6

page 7

page 12

page 14

page 15

page 16

page 17

page 18

research
06/27/2023

Mean Field Theory in Deep Metric Learning

In this paper, we explore the application of mean field theory, a techni...
research
10/04/2022

On Background Bias in Deep Metric Learning

Deep Metric Learning trains a neural network to map input images to a lo...
research
06/09/2021

It Takes Two to Tango: Mixup for Deep Metric Learning

Metric learning involves learning a discriminative representation such t...
research
07/02/2021

Ensemble of Loss Functions to Improve Generalizability of Deep Metric Learning methods

Deep Metric Learning (DML) learns a non-linear semantic embedding from i...
research
11/05/2020

Deep Metric Learning with Spherical Embedding

Deep metric learning has attracted much attention in recent years, due t...
research
12/28/2021

Multi-Head Deep Metric Learning Using Global and Local Representations

Deep Metric Learning (DML) models often require strong local and global ...
research
12/11/2017

Deep metric learning for multi-labelled radiographs

Many radiological studies can reveal the presence of several co-existing...

Please sign up or login with your details

Forgot password? Click here to reset