# GTF: Omnidirectional EPI Transformer for Light Field Super-Resolution

arXiv: https://arxiv.org/abs/2605.04581 · HTML: https://mlsanity.com/paper/2605.04581/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 20 of its 33 references that are in the map.
- First posted on arXiv 6 May 2026.

## Builds on (most cited first, 20 of 20)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Layer Normalization](https://mlsanity.com/paper/1607.06450/) (2016)
- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [SwinIR: Image Restoration Using Swin Transformer](https://mlsanity.com/paper/2108.10257/) (2021)
- [Enhanced Deep Residual Networks for Single Image Super-Resolution](https://mlsanity.com/paper/1707.02921/) (2017)
- [Image Super-Resolution Using Very Deep Residual Channel Attention Networks](https://mlsanity.com/paper/1807.02758/) (2018)
- [Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network](https://mlsanity.com/paper/1609.05158/) (2016)
- [VMamba: Visual State Space Model](https://mlsanity.com/paper/2401.10166/) (2024)
- [Training Region-based Object Detectors with Online Hard Example Mining](https://mlsanity.com/paper/1604.03540/) (2016)
- [Learning-Based View Synthesis for Light Field Cameras](https://mlsanity.com/paper/1609.02974/) (2016)
- [EPINET: A Fully-Convolutional Neural Network Using Epipolar Geometry for Depth from Light Field Images](https://mlsanity.com/paper/1804.02379/) (2018)
- [Disentangling Light Fields for Super-Resolution and Disparity Estimation](https://mlsanity.com/paper/2202.10603/) (2022)
- [Spatial-Angular Interaction for Light Field Image Super-Resolution](https://mlsanity.com/paper/1912.07849/) (2019)
- [Light Field Spatial Super-resolution via Deep Combinatorial Geometry Embedding and Structural Consistency Regularization](https://mlsanity.com/paper/2004.02215/) (2020)
- [Light Field Image Super-Resolution with Transformers](https://mlsanity.com/paper/2108.07597/) (2021)
- [Light Field Image Super-Resolution Using Deformable Convolution](https://mlsanity.com/paper/2007.03535/) (2020)
- [Detail-Preserving Transformer for Light Field Image Super-Resolution](https://mlsanity.com/paper/2201.00346/) (2022)
- [Learning Non-Local Spatial-Angular Correlation for Light Field Image Super-Resolution](https://mlsanity.com/paper/2302.08058/) (2023)
- [NTIRE 2023 Challenge on Light Field Image Super-Resolution: Dataset, Methods and Results](https://mlsanity.com/paper/2304.10415/) (2023)

## Abstract (arXiv)

Light field (LF) image super-resolution benefits from Epipolar Plane Images (EPIs), whose line slopes explicitly encode disparity. However, existing Transformer-based LF SR methods mainly attend to horizontal and vertical EPIs, leaving diagonal epipolar geometry underexplored. We present GTF, an omnidirectional EPI Transformer that explicitly models horizontal, vertical, 45-degree, and 135-degree EPIs within a unified reconstruction framework. GTF combines directional EPI processing, MacPI-based prior injection, adaptive directional fusion, and a topology-preserving feed-forward network to better exploit LF geometry. For the NTIRE 2026 fidelity tracks, we use GTF as the main model, while a lightweight GTF-Tiny variant targets the efficiency track. On five standard LF SR benchmarks covering both real-captured and synthetic scenes, GTF reaches 32.78 dB without inference-time enhancement, and stronger inference settings with EPSW and test-time augmentation further improve performance. Under the NTIRE 2026 efficiency constraint, GTF-Tiny attains 32.57 dB with only 0.915M parameters and 19.81 GFLOPs. In the NTIRE 2026 Light Field Image Super-Resolution Challenge, our submissions rank 3rd on Track 1 and Track 3 and 4th on Track 2. Architecture-evolution, channel-width, and inference analyses further support the effectiveness of diagonal EPI modeling, directional fusion, and the lightweight design.