3D Face Reconstruction Error Decomposed: A Modular Benchmark for Fair and Fast Method Evaluation
This work addresses benchmarking inconsistencies for researchers in 3D face reconstruction, enabling fairer and faster method evaluation, though it is incremental as it builds on existing error measurement components.
The paper tackles the problem of inconsistent and monolithic error measurement in 3D face reconstruction benchmarks by introducing a modular toolkit (M3DFB) that segregates error computation components, revealing that widely used ICP-based estimators perform poorly (correlation as low as 0.41) and that non-rigid alignment improves correlation to over 0.90, with a proposed correction scheme achieving comparable accuracy but running an order of magnitude faster.
Computing the standard benchmark metric for 3D face reconstruction, namely geometric error, requires a number of steps, such as mesh cropping, rigid alignment, or point correspondence. Current benchmark tools are monolithic (they implement a specific combination of these steps), even though there is no consensus on the best way to measure error. We present a toolkit for a Modularized 3D Face reconstruction Benchmark (M3DFB), where the fundamental components of error computation are segregated and interchangeable, allowing one to quantify the effect of each. Furthermore, we propose a new component, namely correction, and present a computationally efficient approach that penalizes for mesh topology inconsistency. Using this toolkit, we test 16 error estimators with 10 reconstruction methods on two real and two synthetic datasets. Critically, the widely used ICP-based estimator provides the worst benchmarking performance, as it significantly alters the true ranking of the top-5 reconstruction methods. Notably, the correlation of ICP with the true error can be as low as 0.41. Moreover, non-rigid alignment leads to significant improvement (correlation larger than 0.90), highlighting the importance of annotating 3D landmarks on datasets. Finally, the proposed correction scheme, together with non-rigid warping, leads to an accuracy on a par with the best non-rigid ICP-based estimators, but runs an order of magnitude faster. Our open-source codebase is designed for researchers to easily compare alternatives for each component, thus helping accelerating progress in benchmarking for 3D face reconstruction and, furthermore, supporting the improvement of learned reconstruction methods, which depend on accurate error estimation for effective training.