DivCon-NeRF: Generating Augmented Rays with Diversity and Consistency for Few-shot View Synthesis

2025-03-17

AI Summary

Existing NeRF methods suffer from overfitting, floating artifacts, and appearance distortions in few-shot novel-view synthesis due to sparse input views. To address this, we propose a dual-spherical ray augmentation strategy: geometrically consistent sampling on a surface sphere and stochastic angular–radial sampling on an inner sphere. We further introduce a consistency-aware filtering mechanism based on probabilistic surface-point ordering, enabling balanced ray diversity and geometric plausibility without requiring precise depth priors. Our approach effectively suppresses rendering artifacts and significantly improves reconstruction robustness. Extensive experiments demonstrate state-of-the-art performance on few-shot novel-view synthesis across the Blender, LLFF, and DTU benchmarks.

Technology Category

• Computer Vision: Computational Photography, Image & Video Synthesis
• Machine Learning: Multi-instance/Multi-view Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Neural Radiance Field (NeRF) has shown remarkable performance in novel view synthesis but requires many multiview images, making it impractical for few-shot scenarios. Ray augmentation was proposed to prevent overfitting for sparse training data by generating additional rays. However, existing methods, which generate augmented rays only near the original rays, produce severe floaters and appearance distortion due to limited viewpoints and inconsistent rays obstructed by nearby obstacles and complex surfaces. To address these problems, we propose DivCon-NeRF, which significantly enhances both diversity and consistency. It employs surface-sphere augmentation, which preserves the distance between the original camera and the predicted surface point. This allows the model to compare the order of high-probability surface points and filter out inconsistent rays easily without requiring the exact depth. By introducing inner-sphere augmentation, DivCon-NeRF randomizes angles and distances for diverse viewpoints, further increasing diversity. Consequently, our method significantly reduces floaters and visual distortions, achieving state-of-the-art performance on the Blender, LLFF, and DTU datasets. Our code will be publicly available.

Problem

• Addresses overfitting in few-shot view synthesis
• Reduces floaters and appearance distortion in NeRF
• Enhances ray diversity and consistency for sparse data

Innovation

• Surface-sphere augmentation preserves camera-surface distance
• Inner-sphere augmentation randomizes angles and distances
• Reduces floaters and distortions in few-shot view synthesis