Event cameras are asynchronous neuromorphic vision sensors with high temporal resolution and no motion blur, offering advantages over standard frame-based cameras especially in high-speed motions and high dynamic range conditions. However, event cameras are unable to capture the overall context of the scene, and produce different events for the same scenery depending on the direction of the motion, creating a challenge in data association. Standard camera, on the other hand, provides frames at a fixed rate that are independent of the motion direction, and are rich in context. In this paper, we present a robust feature tracking method that employs 8-DOF warping model in minimizing the difference between brightness increment patches from events and frames, exploiting the complementary nature of the two data types. Unlike previous works, the proposed method enables tracking of features under complex motions accompanying distortions. Extensive quantitative evaluation over publicly available datasets was performed where our method shows an improvement over state-of-the-art methods in robustness with greatly prolonged feature age and in accuracy for challenging scenarios.

While point cloud semantic segmentation is a significant task in 3D scene understanding, this task demands a time-consuming process of fully annotating labels. To address this problem, recent studies adopt a weakly supervised learning approach under the sparse annotation. Different from the existing studies, this study aims to reduce the epistemic uncertainty measured by the entropy for a precise semantic segmentation. We propose the graphical information gain based attention network called GaIA, which alleviates the entropy of each point based on the reliable information. The graphical information gain discriminates the reliable point by employing relative entropy between target point and its neighborhoods. We further introduce anchor-based additive angular margin loss, ArcPoint. The ArcPoint optimizes the unlabeled points containing high entropy towards semantically similar classes of the labeled points on hypersphere space. Experimental results on S3DIS and ScanNet-v2 datasets demonstrate our framework outperforms the existing weakly supervised methods.

Existing studies on salient object detection (SOD) focus on extracting distinct objects with edge features and aggregating multi-level features to improve SOD performance. However, both performance gain and computational efficiency cannot be achieved, which has motivated us to study the inefficiencies in existing encoder-decoder structures to avoid this trade-off. We propose TRACER which excludes multi-decoder structures and minimizes the learning parameters usage by employing attention guided tracing modules (ATMs), as shown in Fig. 1.

Abstract The performance of document classifiers largely depends on their internal representations of text data. Recent studies have been conducted to identify areas of focus and find latent data spaces to increase the representativeness and the performance of classifiers. In this study, we propose a weight attention layer (WAL) that uses an additional feature of words when computing their attention weights for deep learning models based on attention mechanisms. In the WAL, the attention distribution is calculated through the dot product of the attention weight matrix and a word weight matrix. We utilized information gain, which is one of the feature selection algorithms for the additional feature. To evaluate the proposed method, datasets of helpful reviews, sentiment reviews, and fake reviews were used. These datasets were applied to two deep learning models based on attention mechanisms, including an attention‐based bidirectional long short‐term memory (LSTM) and a hierarchical attention network. As a result of 10‐fold cross validation, the improved performance of the models in terms of accuracy and F1‐score when using WAL is demonstrated.