Light-Field Dataset for Disparity Based Depth Estimation

2025-11-08

AI Summary

Existing light field depth estimation research is hindered by the lack of publicly available datasets that simultaneously exhibit realistic disparity characteristics and controllable geometric parameters. To address this, we introduce the first multimodal light field dataset specifically designed for disparity-based depth estimation, comprising 285 real-world images captured by a Lytro Illum camera and 13 synthetically generated images produced via precise mechanical rig calibration and Blender-based 3D modeling. Leveraging Epipolar Plane Image (EPI) analysis, we quantitatively characterize, for the first time, the influence of camera focal length on the disparity of 3D points—thereby exposing inherent deficiencies in prevailing datasets regarding focal-length sensitivity modeling and occlusion robustness. The dataset is publicly released to support algorithm development, benchmarking, and physically interpretable analysis. It bridges a critical gap in joint real-synthetic validation of light field disparity properties.

Technology Category

• Computer Vision: Computational Photography, Image & Video Synthesis
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Security and Privacy: Data transparency and provenance
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

A Light Field (LF) camera consists of an additional two-dimensional array of micro-lenses placed between the main lens and sensor, compared to a conventional camera. The sensor pixels under each micro-lens receive light from a sub-aperture of the main lens. This enables the image sensor to capture both spatial information and the angular resolution of a scene point. This additional angular information is used to estimate the depth of a 3-D scene. The continuum of virtual viewpoints in light field data enables efficient depth estimation using Epipolar Line Images (EPIs) with robust occlusion handling. However, the trade-off between angular information and spatial information is very critical and depends on the focal position of the camera. To design, develop, implement, and test novel disparity-based light field depth estimation algorithms, the availability of suitable light field image datasets is essential. In this paper, a publicly available light field image dataset is introduced and thoroughly described. We have also demonstrated the effect of focal position on the disparity of a 3-D point as well as the shortcomings of the currently available light field dataset. The proposed dataset contains 285 light field images captured using a Lytro Illum LF camera and 13 synthetic LF images. The proposed dataset also comprises a synthetic dataset with similar disparity characteristics to those of a real light field camera. A real and synthetic stereo light field dataset is also created by using a mechanical gantry system and Blender. The dataset is available at https://github.com/aupendu/light-field-dataset.

Problem

• Addresses the scarcity of suitable light field datasets for depth estimation
• Analyzes the impact of focal position on disparity in 3D scenes
• Provides real and synthetic light field images with disparity characteristics

Innovation

• Light field camera captures spatial and angular information
• Epipolar Line Images enable robust depth estimation
• Dataset combines real and synthetic light field images