Relighting Images in the Wild with a Self-Supervised Siamese Auto-Encoder

被引:6
|
作者
Liu, Yang [1 ]
Neophytou, Alexandros [2 ]
Sengupta, Sunando [2 ]
Sommerlade, Eric [2 ]
机构
[1] Univ Surrey, Guildford, Surrey, England
[2] Microsoft Corp, Reading, Berks, England
关键词
D O I
10.1109/WACV48630.2021.00008
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We propose a self-supervised method for image relighting of single view images in the wild. The method is based on an auto-encoder which deconstructs an image into two separate encodings, relating to the scene illumination and content, respectively. In order to disentangle this embedding information without supervision, we exploit the assumption that some augmentation operations do not affect the image content and only affect the direction of the light. A novel loss function, called spherical harmonic loss, is introduced that forces the illumination embedding to convert to a spherical harmonic vector. We train our model on large-scale datasets such as Youtube 8M and CelebA. Our experiments show that our method can correctly estimate scene illumination and realistically re-light input images, without any supervision or a prior shape model. Compared to supervised methods, our approach has similar performance and avoids common lighting artifacts.
引用
收藏
页码:32 / 40
页数:9
相关论文
共 50 条
  • [1] Self-Supervised Video Hashing With Hierarchical Binary Auto-Encoder
    Song, Jingkuan
    Zhang, Hanwang
    Li, Xiangpeng
    Gao, Lianli
    Wang, Meng
    Hong, Richang
    [J]. IEEE TRANSACTIONS ON IMAGE PROCESSING, 2018, 27 (07) : 3210 - 3221
  • [2] GAEFS: Self-supervised Graph Auto-encoder enhanced Feature Selection
    Tan, Jun
    Gui, Ning
    Qiu, Zhifeng
    [J]. KNOWLEDGE-BASED SYSTEMS, 2024, 290
  • [3] Self-supervised graph clustering via attention auto-encoder with distribution specificity
    Li, Zishi
    Zhu, Changming
    [J]. MULTIMEDIA SYSTEMS, 2024, 30 (03)
  • [4] Modality-Agnostic Self-Supervised Learning with Meta-Learned Masked Auto-Encoder
    Jang, Huiwon
    Tack, Jihoon
    Choi, Daewon
    Jeong, Jongheon
    Shin, Jinwoo
    [J]. ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 36 (NEURIPS 2023), 2023,
  • [5] Context Encoder Guided Self-Supervised Siamese Depth Estimation Based on Stereo Laparoscopic Images
    Li, Wenda
    Hayashi, Yuichiro
    Oda, Masahiro
    Kitasaka, Takayuki
    Misawa, Kazunari
    Mori, Kensaku
    [J]. MEDICAL IMAGING 2021: IMAGE-GUIDED PROCEDURES, ROBOTIC INTERVENTIONS, AND MODELING, 2021, 11598
  • [6] Point-AGM : Attention Guided Masked Auto-Encoder for Joint Self-supervised Learning on Point Clouds
    Liu, Jie
    Yang, Mengna
    Tian, Yu
    Li, Yancui
    Song, Da
    Li, Kang
    Cao, Xin
    [J]. Computer Graphics Forum, 2024, 43 (07)
  • [7] Redundancy Is Not What You Need: An Embedding Fusion Graph Auto-Encoder for Self-Supervised Graph Representation Learning
    Li, Mengran
    Zhang, Yong
    Wang, Shaofan
    Hu, Yongli
    Yin, Baocai
    [J]. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2024, : 1 - 15
  • [8] Discriminative Representation Learning with Supervised Auto-encoder
    Fang Du
    Jiangshe Zhang
    Nannan Ji
    Junying Hu
    Chunxia Zhang
    [J]. Neural Processing Letters, 2019, 49 : 507 - 520
  • [9] Self-supervised Siamese Autoencoders
    Baier, Friederike
    Mair, Sebastian
    Fadel, Samuel G.
    [J]. ADVANCES IN INTELLIGENT DATA ANALYSIS XXII, PT I, IDA 2024, 2024, 14641 : 117 - 128
  • [10] Discriminative Representation Learning with Supervised Auto-encoder
    Du, Fang
    Zhang, Jiangshe
    Ji, Nannan
    Hu, Junying
    Zhang, Chunxia
    [J]. NEURAL PROCESSING LETTERS, 2019, 49 (02) : 507 - 520