No abstract

No abstract

The synthesis of a sustainable material through carbon nitride (C3N4) chemically grafted on waste-derived carbon including carbonizing coals (PM), melamine-urea-formaldehyde resins (MUF-C-1100), and luffa cylindrical sponges (SG), respectively, and its application as sulfur cathode in lithium-sulfur (Li-S) batteries were demonstrated. The Li-S cell assembled by the sulfur (S) cathode with component from C3N4 grafted coal-derived carbon (PM-CN) possesses a specific capacity of 1269.8 mA h g−1 at 0.05 C. At 1 C, the initial specific capacity of PM cathode is only 380.0 mA h g−1, comparable to the PM-CN5 cathode of 681.9 mA h g−1, and PM-CN10 cathode of 580.7 mA h g−1, respectively. And, PM-CN 5 cathode presents the capacity retention of 75.9% with a coulomb efficiency (C.E.) of 97.3% after 200 cycles. The MUF-CN cathode gives a specific capacity of 1335.6 mA h g−1 at 0.05 C, and the capacity retention of 66.7% with a C. E. of 93.6% after 300 cycles at 0.5 C. The SG-CN cathode had a specific capacity of 953.9 mA h g−1 at 0.05 C, and capacity retention of 95.1% with a C. E. of 98.2% after 125 cycles at 1 C. The remarkable improved performances were mainly ascribed to the sustainable materials as S host with micro-meso pore and C3N4 structure providing the strong affinity N sites to lithium polysulfides (LiPSs). This work provides an attractive approach for the preparation of sustainable materials by rational design of grafting C3N4 to waste-derived carbons with functions as S cathode materials for high-performance Li-S batteries.

Visual localization is the task of accurate camera pose estimation within a scene and is a crucial technique for computer vision and robotics. Among the various approaches, relative pose estimation has gained increasing interest because it can generalize to new scenes. This approach learns to regress relative pose between image pairs. However, unreliable regions that contain objects such as the sky, persons, or moving cars are often present in real images, causing noise and interference to localization. In this paper, we propose a novel relative pose estimation pipeline to address the problem. The pipeline features a semantic masking module and an attention module. The two modules help suppress interfering information from unreliable regions, while at the same time emphasizing important features with an attention mechanism. Experiment results show that our framework outperforms alternative methods in the accuracy of camera pose prediction in all scenes.