Annotating tens or hundreds of tiny objects in a given image is laborious yet crucial for a multitude of Computer Vision tasks. Such imagery typically contains objects from various categories, yet the multi-class interactive annotation setting for the detection task has thus far been unex-plored. To address these needs, we propose a novel interactive annotation method for multiple instances of tiny objects from multiple classes, based on a few point-based user in-puts. Our approach, C3Det, relates the full image context with annotator inputs in a local and global manner via late-fusion andfeature-correlation, respectively. We perform ex-periments on the Tiny-DOTA. and LCell datasets using both two-stage and one-stage object detection architectures to verify the efficacy of our approach. Our approach outper-forms existing approaches in interactive annotation, achieving higher mAP with fewer clicks. Furthermore, we validate the annotation efficiency of our approach in a user study where it is shown to be 2.85x faster and yield only 0.36x task load (NASA-TLX, lower is better) compared to manual annotation. The code is available at https://github.com/ChungYi347/Interactive-Multi-Class-Tiny-Object-Detection.

To achieve high spectral efficiency, a modern cellular network such as LTE or 5G New Radio (NR) aims to operate with full frequency reuse. This deployment will significantly increase the level of Co-Channel Interference (CCI) for cell-edge User Equipments (UEs), and the CCI has become a major throughput-limiting factor. Thus, the suppression of CCI in the 5G network is the most important feature to increase downlink throughput. In order to mitigate CCI, Interference Whitening (IW) is an effective low-complexity linear method to suppress colored interference in a MIMO-OFDM system. However, conventional IW can degrade the performance when the noise-dominant environment due to limited samples, e.g., DMRS (De-Modulation Reference Signal). To address that, we propose a Reinforcement Learning based Interference Whitening (RL-IW) that adaptively controls the IW mode by learning algorithm. The experimental results show that RL-IW has performance gain in terms of both BLER (Block Error Rate) and downlink throughput than conventional IW.

This paper presents an image-based rendering (IBR) system based on RGB-D images. The input of our system consists of RGB-D images captured at sparse locations in the scene and can be expanded by adding new RGB-D images. The sparsity of RGB-D images increases the usability of our system as the user need not capture a RGB-D image stream in a single shot, which may require careful planning for a hand-held camera. Our system begins with a single RGB-D image and images are incrementally added one by one. For each newly added image, a batch process is performed to align it with previously added images. The process does not include a global alignment step, such as bundle adjustment, and can be completed quickly by computing only local alignments of RGB-D images. Aligned images are represented as a graph, where each node is an input image and an edge contains relative pose information between nodes. A novel view image is rendered by picking the nearest input as the reference image and then blending the neighboring images based on depth information in real time. Experimental results with indoor scenes using Microsoft Kinect demonstrate that our system can synthesize high quality novel view images from a sparse set of RGB-D images.