Shadow characteristics are of great importance for scene understanding. Existing works mainly consider shadow regions as binary masks, often leading to imprecise detection results and suboptimal performance for scene understanding. We demonstrate that such an assumption oversimplifies light-object interactions in the scene, as the scene details under either hard or soft shadows remain visible to a certain degree. Based on this insight, we aim to reformulate the shadow detection paradigm from the opacity perspective, and introduce a new fine-grained shadow detection method. In particular, given an input image, we first propose a shadow opacity augmentation module to generate realistic images with varied shadow opacities. We then introduce a shadow feature separation module to learn the shadow position and opacity representations separately, followed by an opacity mask prediction module that fuses these representations and predicts fine-grained shadow detection results. In addition, we construct a new dataset with opacity-annotated shadow masks across varied scenarios. Extensive experiments demonstrate that our method outperforms the baselines qualitatively and quantitatively, enhancing a wide range of applications, including shadow removal, shadow editing, and 3D reconstruction.

In this section we provide a more detailed derivation of the proposed loss function (Equation 17). We make use of the fact that the negative entropy of the Dirichlet distribution is equivalent to the reverse KL-divergence to a flat Dirichlet, up to an additive constant which doesn't depend on the model. Additionally, we can see that by adding +1 to the target concentration parameters ˆ, we are now minimizing an upper bound to the KL-divergence between the mean and the ensemble. Then we divide through by ˆ 0 and drop the additive constant. This yields a loss which is remarkable similar to an ELBO.

Appendix The appendix is organized as follows. We also provide details of the annotation/calibration process and the baseline neural networks (NNs) in Section D and E, respectively. We discuss results regarding each weather condition and consideration of the K-Radar dataset as a pre-training dataset for other Radar tensor datasets in Section F and G, respectively. Finally, we introduce details of devkits and list relevant URLs to help with understanding the content of the paper in Section H and I, respectively. A.1 Additional samples of the K-Radar dataset and explanation of LPCs for each weather condition In the sleet (Figure 8-(e)) or heavy snow (Figure 8-(g)) condition, the Lidar point cloud (LPC) measurements of some objects ahead are lost when the ego-vehicle is driving.

Unmanned Aerial V ehicles (UA Vs), equipped with cameras, are employed in numerous applications, including aerial photography, surveillance, and agriculture. In these applications, robust object detection and tracking are essential for the effective deployment of UA Vs.