Skip to Main content Skip to Navigation
Theses

Learnable factored image representation for visual discovery

Abstract : This thesis proposes an approach for analyzing unpaired visual data annotated with time stamps by generating how images would have looked like if they were from different times. To isolate and transfer time dependent appearance variations, we introduce a new trainable bilinear factor separation module. We analyze its relation to classical factored representations and concatenation-based auto-encoders. We demonstrate this new module has clear advantages compared to standard concatenation when used in a bottleneck encoder-decoder convolutional neural network architecture. We also show that it can be inserted in a recent adversarial image translation architecture, enabling the image transformation to multiple different target time periods using a single network.
Complete list of metadatas

Cited literature [71 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-02931611
Contributor : Abes Star :  Contact
Submitted on : Monday, September 7, 2020 - 9:48:16 AM
Last modification on : Wednesday, September 23, 2020 - 6:12:25 AM

File

Dalens-2019-These.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-02931611, version 1

Collections

Citation

Théophile Dalens. Learnable factored image representation for visual discovery. Computer Vision and Pattern Recognition [cs.CV]. Université Paris sciences et lettres, 2019. English. ⟨NNT : 2019PSLEE036⟩. ⟨tel-02931611⟩

Share

Metrics

Record views

61

Files downloads

14