Search papers, labs, and topics across Lattice.
The paper introduces DF3DV-1K, a comprehensive dataset designed for benchmarking distractor-free novel view synthesis, consisting of 1,048 scenes with both clean and cluttered images. This dataset addresses a critical gap in the existing literature by providing a large-scale resource that includes 89,924 images and spans various environments and distractor types. The authors benchmark nine recent distractor-free radiance field methods, revealing significant performance improvements when employing a diffusion-based 2D enhancer, with notable gains in PSNR and LPIPS metrics.
Benchmarking reveals that a diffusion-based 2D enhancer can significantly elevate the performance of distractor-free radiance field methods, achieving nearly 1 dB PSNR improvement.
Advances in radiance fields have enabled photorealistic novel view synthesis. In several domains, large-scale real-world datasets have been developed to support comprehensive benchmarking and to facilitate progress beyond scene-specific reconstruction. However, for distractor-free radiance fields, a large-scale dataset with clean and cluttered images per scene remains lacking, limiting the development. To address this gap, we introduce DF3DV-1K, a large-scale real-world dataset comprising 1,048 scenes, each providing clean and cluttered image sets for benchmarking. In total, the dataset contains 89,924 images captured using consumer cameras to mimic casual capture, spanning 128 distractor types and 161 scene themes across indoor and outdoor environments. A curated subset of 41 scenes, DF3DV-41, is systematically designed to evaluate the robustness of distractor-free radiance field methods under challenging scenarios. Using DF3DV-1K, we benchmark nine recent distractor-free radiance field methods and 3D Gaussian Splatting, identifying the most robust methods and the most challenging scenarios. Beyond benchmarking, we demonstrate an application of DF3DV-1K by fine-tuning a diffusion-based 2D enhancer to improve radiance field methods, achieving average improvements of 0.96 dB PSNR and 0.057 LPIPS on the held-out set (e.g., DF3DV-41) and the On-the-go dataset. We hope DF3DV-1K facilitates the development of distractor-free vision and promotes progress beyond scene-specific approaches.
