Track trials, findings, and clinical evidence as it lands.
Search papers, labs, and topics across Lattice.
To overcome systematic stereo matching failures on flat, textureless objects under single-source illumination, the authors formulate Specular Reflection Disparity Estimation (SRDE), which leverages the precise reflection geometry of specular highlights instead of photometric surface textures. The method acts as a modular post-processing step that selectively overrides predictions in specular regions without requiring retraining of underlying neural disparity networks. On synthetic benchmarks, this geometric formulation reduces endpoint error (EPE) by more than 52% while demonstrating strong zero-shot transfer to real-world scenes.
Specular highlights usually break stereo vision, but treating their reflection geometry as an explicit depth cue slashes disparity error by over 52% on textureless surfaces without retraining the underlying neural network.
Multicamera imaging and camera arrays have become ubiquitous in many applications, such as autonomous driving, robot control, or virtual reality, and accurate disparity maps of objects and their environment are essential for reliable operation. Despite recent advances in neural networks, correctly estimating the disparity of flat and textureless objects remains challenging. In particular, we consider a scenario defined by flat, textureless surfaces illuminated by a single fixed light source, resulting in one dominant specular reflection visible on the object surface. Under these conditions, reliable geometric and photometric cues are missing, and the specular reflection often causes mispredictions, especially when using conventional methods that rely on texture information to match corresponding pixels. To address this issue, the novel specular reflection disparity estimation (SRDE) algorithm is introduced, which is specifically designed for the constrained scenario of planar, textureless objects and single-source illumination. Unlike conventional stereo matching methods, SRDE ignores texture and instead leverages the geometric properties of specular reflections by incorporating the position information of the reflective region, the light source, and the camera setup. We show that SRDE outperforms existing methods by a notable margin, achieving more than a 52 % improvement in endpoint error (EPE) on synthetic images. Further tests demonstrate superior performance on real-world data. Furthermore, we integrate SRDE into existing neural disparity estimation pipelines by selectively replacing predictions in specular regions without modifying the backbone model. This hybrid strategy enables additional performance gains without requiring network retraining.