1 March 2014 in AIP Advances
The current focus in the field of life science is the use of tissue engineering scaffolds to repair human organs, which has shown great potential in clinical applications. Extracellular matrix morphology and the performance and internal structure of natural organs are required to meet certain requirements. Therefore, integrating multiple processes can effectively overcome the limitations of the individual processes and can take into account the needs of scaffolds for the material, structure, mechanical properties and many other aspects. This study combined the biological 3D printing technology and the near-field electro-spinning (NFES) process to prepare a multi-scale controlled tissue engineering scaffold. While using 3D printing technology to directly prepare the macro-scaffold, the compositing NFES process to build tissue micro-morphology ultimately formed a tissue engineering scaffold which has the specific extracellular matrix structure. This scaffold not only takes into account the material, structure, performance and many other requirements, but also focuses on resolving the controllability problems in macro- and micro-forming which further aim to induce cell directed differentiation, reproduction and, ultimately, the formation of target tissue organs. It has in-depth immeasurable significance to build ideal scaffolds and further promote the application of tissue engineering.
15 June 2024 in Neural Networks
11 August 2023 in IEEE Signal Processing Letters
Video frame interpolation aims to improve the video quality by increasing the frame rate. Existing methods adopt the cascaded architecture. They first estimate intermediate flow maps and then refine the synthesized intermediate frames with contextual features separately. However, the separated flow estimation and refined module ignore the mutual facilitation of them in frame interpolation. Following this issue, we propose a Dual-Stream Fused Network (DSF-Net) to joint flow estimation and refinement module for frame interpolation. Specifically, it first extracts the contextual features from input frames by a contextual feature extractor module, and then jointly refines the intermediate flow maps with the extracted features through a coarse-to-fine frame synthesis module. DSF-Net allows the intermediate flow and the contextual features to benefit each other while generating sharper moving objects and capturing better textual details. Experimental results demonstrate that DSF-Net performs consistently better than existing SOTA methods.
17 April 2026 in IEEE Transactions on Geoscience and Remote Sensing
Unsupervised domain adaptive object detection in remote sensing images aims to transfer knowledge from a labeled source domain to an unlabeled target domain, mitigating performance degradation caused by domain shift. Existing methods commonly adopt a teacher–student framework, where a single teacher, updated via an exponential moving average (EMA) of the student parameters to generate pseudo-labels for training. While this framework effectively transfers knowledge from the teacher to the student, the quality of pseudo-labels and the alignment of source and target domain features remain key challenges for reducing domain shift. To address these challenges, a dual-teacher prototype consistency (DTPC) method is proposed. Specifically, DTPC employs two available object detection models as teachers to provide complementary guidance for the student. The larger teacher, equipped with a deeper backbone and kept frozen, extracts domain-invariant prototype anchors through a class-aware prototype update mechanism, thereby facilitating effective feature-level alignment between the source and target domains. Furthermore, the larger teacher and an EMA-updated smaller teacher jointly generate pseudo-labels to supervise the student with soft and hard label losses, improving the quality and reliability of pseudo-labels. Extensive experiments on two remote sensing benchmarks demonstrate that DTPC achieves the state-of-the-art (SOTA) results, surpassing the second-best methods by 4.6% and 4.3% mean average precision (mAP) on the NWPU VHR-10 → DIOR and VisDrone → DroneVehicle adaptation settings, respectively.
10 August 2026 in Information Fusion