About me
I am a second-year master’s student at the University of Science and Technology, Beijing, majoring in Optical Fiber Sensing. However, I have a strong passion for AI-related content, particularly in 2D and 3D AIGC. I am especially fascinated by 3D mesh and digital human-related technologies. In 2023, I briefly studied Diffusion and then fine-tuned LoRA for Stable Diffusion. As for the results, let’s just say they were… not presentable, haha!
In 2024, I began exploring 3D AIGC and digital human-related technologies. This field involves a wide range of technologies, and most of my time has been spent reproducing open-source algorithms. Of course, I haven’t been modifying models or datasets and retraining them from scratch, as I don’t have access to sufficient GPU resources, nor do I have a mentor to guide me through the process.
Up to now, I have explored various fields: text-to-motion, motion capture, music-to-dance, interaction with scenes, automatic speech recognition (ASR), text-to-speech (TTS), singing voice conversion (SVC), music generation, 3D avatar animation, auto-rigging, mesh generation, eye tracking, and gesture recognition…… I don’t consider myself a dedicated algorithm developer, but fortunately, I possess strong engineering skills, which make it relatively easy for me to reproduce and implement open-source algorithms.
I think this outcome is quite impressive—using music to generate 3D dance motions (SMPL data), then converting them into VMD animations (or alternatively, FBX or BVH formats), and finally driving the PMX model to animate. I have also published it on other platforms: rednotes, bilibili.
You can find my CV here: Ruiqing Tang’s Curriculum Vitae.
