ABSTRACT
This article presents a new approach for surface normal recovery from polarization images under an unknown distant light. Polarization provides rich cues of object geometry and material, but it is also influenced by different lighting conditions. Different from previous Shape-from-Polarization (SfP) methods, which rely on handcrafted or data-driven priors, we analytically investigate the benefits of estimating distant lighting for resolving the ambiguity in normal estimation from SfP using the polarimetric Bidirectional Reflectance Distribution Function (pBRDF) based image formation model. We then propose a two-stage learning framework that first effectively exploits polarization and shading cues to estimate the reflectance and lighting information and then optimizes the initial normal as the geometric prior. Leveraging the normal prior with the polarization cues from the input images, our network further generates the surface normal with more details in the second stage. We also present a data generation pipeline derived from the pBRDF model enabling model training and create a real dataset for evaluation of SfP approaches. Extensive ablation studies show the effectiveness of our designed architecture, and our approach outperforms existing methods in quantitative and qualitative experiments on real data.
ABSTRACT
In this article, we investigate the problem of panoramic image reflection removal to relieve the content ambiguity between the reflection layer and the transmission scene. Although a partial view of the reflection scene is attainable in the panoramic image and provides additional information for reflection removal, it is not trivial to directly apply this for getting rid of undesired reflections due to its misalignment with the reflection-contaminated image. We propose an end-to-end framework to tackle this problem. By resolving misalignment issues with adaptive modules, the high-fidelity recovery of reflection layer and transmission scenes is accomplished. We further propose a new data generation approach that considers the physics-based formation model of mixture images and the in-camera dynamic range clipping to diminish the domain gap between synthetic and real data. Experimental results demonstrate the effectiveness of the proposed method and its applicability for mobile devices and industrial applications.
ABSTRACT
Soil contamination with heavy metals is a relatively serious issue in China. Traditional soil heavy metal survey methods cannot meet the demand for rapid and real-time large-scale area soil heavy metal surveys. We chose a typical mining area in Henan Province as the study area, collected 124 soil samples in the field and obtained their soil hyperspectral data indoors using a spectrometer. After different spectral transformations of the soil spectral curves, Pearson correlation coefficients (PCC) between them and the heavy metals Cd, Cr, Cu, and Ni were calculated, and after correlation evaluation, the best spectral transformations for each heavy metal were determined and preselected characteristic wavebands were obtained. Then the support vector machine recursive feature elimination cross-validation (SVM-RFECV) is used to select among the preselected feature wavebands to obtain the final modeled wavebands, and the Adaptive Boosting (AdaBoost), Gradient Boosting Decision Tree (GBDT), Random Forest (RF), and Partial Least Squares (PLS) methods were used to establish the inversion model. The results showed that the PCC-SVM-RFECV can effectively select characteristic wavebands with high contribution to modeling from high-dimensional data. Spectral transformations methods can improve the correlation of spectra with heavy metals. The location and quantity of characteristic wavebands for the four heavy metals were different. The accuracy of AdaBoost was significantly better than that of GBDT, RF, and PLS (i.e., Ni: [Formula: see text]). This study can provide a technical reference for the use of hyperspectral inversion models for large-scale monitoring of soil heavy metal content.
