Multi-modal sensor fusion in Bird's Eye View (BEV) representation has become the leading approach for 3D object detection. However, existing methods often rely on depth estimators or transformer encoders to transform image features into BEV space, which reduces robustness or introduces significant computational overhead. Moreover, the insufficient geometric guidance in view transformation results in ray-directional misalignments, limiting the effectiveness of BEV representations. To address these challenges, we propose Efficient View Transformation (EVT), a novel 3D object detection framework that constructs a well-structured BEV representation, improving both accuracy and efficiency. Our approach focuses on two key aspects. First, Adaptive Sampling and Adaptive Projection (ASAP), which utilizes LiDAR guidance to generate 3D sampling points and adaptive kernels, enables more effective transformation of image features into BEV space and a refined BEV representation. Second, an improved query-based detection framework, incorporating group-wise mixed query selection and geometry-aware cross-attention, effectively captures both the common properties and the geometric structure of objects in the transformer decoder. On the nuScenes test set, EVT achieves state-of-the-art performance of 75.3% NDS with real-time inference speed.

The traffic speed on a given road segment is affected by the current and past speeds on nearby segments, and the influence further cascades into the rest of a transport network. Thus, a successful forecasting model should consider not only the impact of neighboring road segments but also that of distant segments. Based on this principle, the approach proposed here projects the topology of a real traffic network into the structure of a deep neural network in order to accommodate citywide spatial correlations as well as temporal dependencies. This approach leads to interesting model interpretations in terms of traffic state transition and propagation, which form a basis for extending the proposed forecasting model. The present study was conducted with a large-scale data set collected over 10 months, and traffic speeds were successfully forecasted for 170 road segments in Gangnam, Seoul, Korea.

This paper addresses an information-theoretic aspect of k-means and spectral clustering. First, we revisit the k-means clustering and show that its objective function is approximately derived from the minimum entropy principle when the Renyi's quadratic entropy is used. Then we present a maximum within-clustering association that is derived using a quadratic distance measure in the framework of minimum entropy principle, which is very similar to a class of spectral clustering algorithms that is based on the eigen-decomposition method.