arXiv:2511.20804v1 Announce Type: cross 
Abstract: Neural Radiance Fields (NeRFs) have demonstrated remarkable capabilities in 3D reconstruction and novel view synthesis. However, most existing NeRF frameworks require complete retraining when new views are introduced incrementally, limiting their applicability in domains where data arrives sequentially. This limitation is particularly problematic in satellite-based terrain analysis, where regions are repeatedly observed over time. Incremental refinement of NeRFs remains underexplored, and naive approaches suffer from catastrophic forgetting when past data is unavailable. We propose $\Delta$-NeRF, a unique modular residual framework for incremental NeRF refinement. $\Delta$-NeRF introduces several novel techniques including: (1) a residual controller that injects per-layer corrections into a frozen base NeRF, enabling refinement without access to past data; (2) an uncertainty-aware gating mechanism that prevents overcorrection by adaptively combining base and refined predictions; and (3) a view selection strategy that reduces training data by up to 47\% while maintaining performance. Additionally, we employ knowledge distillation to compress the enhanced model into a compact student network (20\% of original size). Experiments on satellite imagery demonstrate that $\Delta$-NeRF achieves performance comparable to joint training while reducing training time by 30-42\%. $\Delta$-NeRF consistently outperforms existing baselines, achieving an improvement of up to 43.5\% in PSNR over naive fine-tuning and surpassing joint training on some metrics.

يقدم $	ext{Δ}$-NeRF نهجًا مبتكرًا لتحسين مجالات الإشعاع العصبي (NeRF) بشكل تدريجي، حيث يتناول تحدي إعادة التدريب عند إضافة وجهات جديدة. يسمح هذا الإطار بتحسينات في إعادة بناء ثلاثية الأبعاد وتوليد وجهات جديدة، مما يكون مفيدًا بشكل خاص في التطبيقات مثل تحليل التضاريس المعتمد على الأقمار الصناعية حيث يتم جمع البيانات على مر الزمن.

La introducción de $	ext{Δ}$-NeRF presenta un enfoque novedoso para el refinamiento incremental de los Campos de Radiancia Neurales (NeRF), abordando el desafío de la reentrenamiento al agregar nuevas vistas. Este marco permite mejoras en la reconstrucción 3D y la síntesis de nuevas vistas, siendo particularmente beneficioso en aplicaciones como el análisis de terrenos basado en satélites, donde los datos se recopilan a lo largo del tiempo.

L'introduction de $	ext{Δ}$-NeRF propose une approche novatrice pour le raffinement incrémental des Neural Radiance Fields (NeRF), répondant au défi de la réentraînement lors de l'ajout de nouvelles vues. Ce cadre permet des améliorations dans la reconstruction 3D et la synthèse de nouvelles vues, particulièrement bénéfique dans des applications telles que l'analyse de terrain par satellite où les données sont collectées au fil du temps.

The introduction of $	ext{Δ}$-NeRF presents a novel approach for the incremental refinement of Neural Radiance Fields (NeRFs), addressing the challenge of retraining when new views are added. This framework allows for enhancements in 3D reconstruction and novel view synthesis, particularly beneficial in applications like satellite-based terrain analysis where data is collected over time.

$\Delta$-NeRF: Incremental Refinement of Neural Radiance Fields through Residual Control and Knowledge Transfer

arXiv:2501.13104v2 Announce Type: replace 
Abstract: Neural Radiance Fields (NeRFs) have remodeled 3D scene representation since release. NeRFs can effectively reconstruct complex 3D scenes from 2D images, advancing different fields and applications such as scene understanding, 3D content generation, and robotics. Despite significant research progress, a thorough review of recent innovations, applications, and challenges is lacking. This survey compiles key theoretical advancements and alternative representations and investigates emerging challenges. It further explores applications on reconstruction, highlights NeRFs' impact on computer vision and robotics, and reviews essential datasets and toolkits. By identifying gaps in the literature, this survey discusses open challenges and offers directions for future research.

لقد حولت حقول الإشعاع العصبي (NeRF) تمثيل المشاهد ثلاثية الأبعاد، مما يتيح إعادة بناء بيئات معقدة من صور ثنائية الأبعاد. تسلط دراسة حديثة الضوء على التقدمات والتطبيقات والتحديات المرتبطة بـ NeRF، مما يبرز أهميتها في مجالات مثل رؤية الكمبيوتر والروبوتات.

Los Campos de Radiancia Neuronal (NeRF) han transformado la representación de escenas 3D, permitiendo la reconstrucción de entornos complejos a partir de imágenes 2D. Una reciente encuesta destaca los avances, aplicaciones y desafíos asociados con los NeRF, enfatizando su importancia en campos como la visión por computadora y la robótica.

Les champs de radiance neuronaux (NeRF) ont transformé la représentation des scènes 3D, permettant la reconstruction d'environnements complexes à partir d'images 2D. Une récente enquête met en lumière les avancées, les applications et les défis associés aux NeRF, soulignant leur importance dans des domaines tels que la vision par ordinateur et la robotique.

Neural Radiance Fields (NeRFs) have transformed the representation of 3D scenes, enabling the reconstruction of complex environments from 2D images. A recent survey highlights the advancements, applications, and challenges associated with NeRFs, emphasizing their significance in fields such as computer vision and robotics.

Neural Radiance Fields for the Real World: A Survey

arXiv:2512.06438v1 Announce Type: new 
Abstract: The generation of high-fidelity, animatable 3D human avatars remains a core challenge in computer graphics and vision, with applications in VR, telepresence, and entertainment. Existing approaches based on implicit representations like NeRFs suffer from slow rendering and dynamic inconsistencies, while 3D Gaussian Splatting (3DGS) methods are typically limited to static head generation, lacking dynamic control. We bridge this gap by introducing AGORA, a novel framework that extends 3DGS within a generative adversarial network to produce animatable avatars. Our key contribution is a lightweight, FLAME-conditioned deformation branch that predicts per-Gaussian residuals, enabling identity-preserving, fine-grained expression control while allowing real-time inference. Expression fidelity is enforced via a dual-discriminator training scheme leveraging synthetic renderings of the parametric mesh. AGORA generates avatars that are not only visually realistic but also precisely controllable. Quantitatively, we outperform state-of-the-art NeRF-based methods on expression accuracy while rendering at 250+ FPS on a single GPU, and, notably, at $\sim$9 FPS under CPU-only inference - representing, to our knowledge, the first demonstration of practical CPU-only animatable 3DGS avatar synthesis. This work represents a significant step toward practical, high-performance digital humans. Project website: https://ramazan793.github.io/AGORA/

تم تقديم AGORA كإطار جديد يعزز من توليد الصور الرمزية البشرية ثلاثية الأبعاد القابلة للتحريك من خلال توسيع تقنية 3D Gaussian Splatting ضمن شبكة خصومة توليدية. يعالج هذا التطور قيود الأساليب الحالية، مثل بطء العرض ونقص التحكم الديناميكي، مما يسمح بالاستدلال في الوقت الحقيقي والتحكم الدقيق في التعبيرات.

AGORA se ha presentado como un nuevo marco que mejora la generación de avatares humanos 3D animables al extender el 3D Gaussian Splatting dentro de una red generativa adversarial. Este desarrollo aborda las limitaciones de los métodos existentes, como el renderizado lento y la falta de control dinámico, permitiendo la inferencia en tiempo real y un control de expresión detallado.

AGORA a été introduit comme un cadre novateur qui améliore la génération d'avatars humains 3D animables en étendant le 3D Gaussian Splatting au sein d'un réseau antagoniste génératif. Ce développement répond aux limitations des méthodes existantes, telles que le rendu lent et le manque de contrôle dynamique, permettant une inférence en temps réel et un contrôle d'expression précis.

AGORA has been introduced as a novel framework that enhances the generation of animatable 3D human avatars by extending 3D Gaussian Splatting within a generative adversarial network. This development addresses the limitations of existing methods, such as slow rendering and lack of dynamic control, enabling real-time inference and fine-grained expression control.

$\Delta$-NeRF: Incremental Refinement of Neural Radiance Fields through Residual Control and Knowledge Transfer

Was this article worth reading? Share it

LucidQuery AI

Attentive AI

MyArchitectAI