arXiv subjects

Shiguang Zhang

Publications and source records attributed to Shiguang Zhang.

3 recordsLinked to original sources

arXiv2024

Zero-Shot Object Goal Visual Navigation With Class-Independent Relationship Network

This paper investigates the zero-shot object goal visual navigation problem. In the object goal visual navigation task, the agent needs to locate navigation targets from its egocentric visual input. "Zero-shot" means that the target the agent needs to find is not trained during the training phase. To address the issue of coupling navigation ability with target features during training, we propose the Class-Independent Relationship Network (CIRN). This method combines target detection information with the relative semantic similarity between the target and the navigation target, and constructs a brand new state representation based on similarity ranking, this state representation does not include target feature or environment feature, effectively decoupling the agent's navigation ability from target features. And a Graph Convolutional Network (GCN) is employed to learn the relationships between different objects based on their similarities. During testing, our approach demonstrates strong generalization capabilities, including zero-shot navigation tasks with different targets and environments. Through extensive experiments in the AI2-THOR virtual environment, our method outperforms the current state-of-the-art approaches in the zero-shot object goal visual navigation task. Furthermore, we conducted experiments in more challenging cross-target and cross-scene settings, which further validate the robustness and generalization ability of our method. Our code is available at: https://github.com/SmartAndCleverRobot/ICRA-CIRN.

cs.CV↗

arXiv2023

Robust 3.7 V-Na$_{2/3}$[Cu$_{1/3}$Mn$_{2/3}$]O$_2$ Cathode for Na-ion Batteries

Na-ion batteries (NIBs), which are recognized as a next-generation alternative technology for energy storage, still suffer from commercialization constraints due to the lack of low-cost, high-performance cathode materials. Since our first discovery of Cu$^{3+}$/Cu$^{2+}$ electrochemistry in 2014, numerous Cu-substituted/doped materials have been designed for NIBs. However for almost ten years, the potential of Cu$^{3+}$/Cu$^{2+}$ electrochemistry has been grossly underappreciated and normally regarded as a semielectrochemically active redox. Here, we re-synthesized P2-Na$_{2/3}$[Cu$_{1/3}$Mn$_{2/3}$]O$_2$ and reinterpreted it as a high-voltage, cost-efficient, air-stable, long-life, and high-rate cathode material for NIBs, which demonstrates a high operating voltage of 3.7 V and a completely active Cu$^{3+}$/Cu$^{2+}$ redox reaction. The 2.3 Ah cylindrical cells exhibit excellent cycling (93.1% capacity after 2000 cycles), high rate (97.2% capacity at 10C rate), good low-temperature performance (86.6% capacity at -30$^\circ$C), and high safety, based on which, a 56 V-11.5 Ah battery pack for E-bikes is successfully constructed, exhibiting stable cycling (96.5% capacity at the 800th cycle) and a long driving distance (36 km, tester weight 65 kg). This work offers a commercially feasible cathode material for low-cost, high-voltage NIBs, paving the way for advanced NIBs in power and stationary energy storage applications.

cond-mat.mtrl-sci↗

arXiv2022

Learning Dual Memory Dictionaries for Blind Face Restoration

To improve the performance of blind face restoration, recent works mainly treat the two aspects, i.e., generic and specific restoration, separately. In particular, generic restoration attempts to restore the results through general facial structure prior, while on the one hand, cannot generalize to real-world degraded observations due to the limited capability of direct CNNs' mappings in learning blind restoration, and on the other hand, fails to exploit the identity-specific details. On the contrary, specific restoration aims to incorporate the identity features from the reference of the same identity, in which the requirement of proper reference severely limits the application scenarios. Generally, it is a challenging and intractable task to improve the photo-realistic performance of blind restoration and adaptively handle the generic and specific restoration scenarios with a single unified model. Instead of implicitly learning the mapping from a low-quality image to its high-quality counterpart, this paper suggests a DMDNet by explicitly memorizing the generic and specific features through dual dictionaries. First, the generic dictionary learns the general facial priors from high-quality images of any identity, while the specific dictionary stores the identity-belonging features for each person individually. Second, to handle the degraded input with or without specific reference, dictionary transform module is suggested to read the relevant details from the dual dictionaries which are subsequently fused into the input features. Finally, multi-scale dictionaries are leveraged to benefit the coarse-to-fine restoration. Moreover, a new high-quality dataset, termed CelebRef-HQ, is constructed to promote the exploration of specific face restoration in the high-resolution space.

cs.CV↗