2K Retrofit: Entropy-Guided Efficient Sparse Refinement for High-Resolution 3D Geometry Prediction

TL;DR

2K Retrofit is a novel framework that enables efficient high-resolution 3D geometry prediction at 2K resolution by selectively refining uncertain regions, significantly improving accuracy and speed without retraining models.

Contribution

It introduces an entropy-guided sparse refinement method that allows any geometric foundation model to perform high-resolution inference efficiently without retraining.

Findings

Achieves state-of-the-art accuracy on benchmark datasets.

Reduces computational overhead for 2K resolution inference.

Maintains high fidelity in high-uncertainty regions.

Abstract

High-resolution geometric prediction is essential for robust perception in autonomous driving, robotics, and AR/MR, but current foundation models are fundamentally limited by their scalability to real-world, high-resolution scenarios. Direct inference on 2K images with these models incurs prohibitive computational and memory demands, making practical deployment challenging. To tackle the issue, we present 2K Retrofit, a novel framework that enables efficient 2K-resolution inference for any geometric foundation model, without modifying or retraining the backbone. Our approach leverages fast coarse predictions and an entropy-based sparse refinement to selectively enhance high-uncertainty regions, achieving precise and high-fidelity 2K outputs with minimal overhead. Extensive experiments on widely used benchmark demonstrate that 2K Retrofit consistently achieves state-of-the-art accuracy and speed, bridging the gap between research advances and scalable deployment in high-resolution 3D vision applications. Code will be released upon acceptance.