Predicting topological defects has traditionally required slow, resource-intensive simulations—but that is all starting to change. Researchers at Chungnam National University are looking to solve this problem, with a deep learning method that predicts stable defect configurations in nematic liquid crystals in milliseconds rather than hours. In nematic liquid crystals, molecules can rotate freely while remaining roughly aligned. Now, researchers led by Professor Jun-Hee Na from Chungnam National University, Republic of Korea, have developed a faster way to predict stable defect configurations using deep learning, replacing time-consuming conventional numerical simulations. The model employs 3D U-Net architecture, a convolutional neural network widely used in scientific and medical image analysis, to capture both global orientational order and local defect structures. The framework works by directly linking prescribed boundary conditions to the final equilibrium structure. Boundary information is fed into the neural network, which then predicts the complete molecular alignment field, including defect locations and shapes. The model was trained on data generated using conventional simulations covering a wide range of alignment patterns. Once trained, it can accurately predict new configurations it has never seen, with results that agree closely with both simulations and experiments. Here is how this can help: Speed the design of advanced materials that currently rely on lengthy trial-and-error processes. Provide a clear and controllable platform for observing how defects form, move, and reorganize. Reduce simulation times from hours to milliseconds. Looking to the future, we are going to continue to see new research in this area, opening up new possibilities for designing materials with specific defect architectures for optical devices and…

What if we could use AI (artificial intelligence) and 3D forest reconstruction from remote sensing data to help find and reconstruct trees in a forest? Researchers at Purdue University’s Department of Computer Science and Institute for Digital Forestry and Germany’s Kiel University are doing exactly this, leveraging AI to isolate and reconstruct forest trees. Up until now, existing algorithms could only partially reconstruct the shape of a single tree from a clean point-cloud dataset acquired by laser-scanning technologies. Now, researchers have introduced TreeStructor in IEEE transactions on geoscience and remote sensing. Lidar (light detection and ranging) works by shooting laser pulses at the target objects, then detecting the reflected light. Tree trunks and branches standing behind the reflecting objects remain invisible, and the canopy dissipates the reflections to almost random directions. The workaround is to combine the results of multiple scans from various angles from the ground and sometimes from a drone flying above. Here is how this can help: Detect and isolate repeating parts and capture tree shapes. Provide scientific research. Could lead to economic benefits in the future. Looking to the future, this could open new opportunities for digital twins, forest reconstruction, and more, as we look to create images of shapes. Urban structures, furniture, cars and other human-built products display a high degree of symmetry, making them easier to detect from point-cloud datasets collected by lidar and other remote-sensing technology. But now we may have technology to help create similar datasets for…