In this work, we investigate a simple and must-known conditional generative framework based on Vector Quantised-Variational AutoEncoder (VQ-VAE) and Generative Pre-trained Transformer (GPT) for human motion generation from textural descriptions. We show that a simple CNN-based VQ-VAE with commonly used training recipes (EMA and Code Reset) allows us to obtain high-quality discrete representations. For GPT, we incorporate a simple corruption strategy during the training to alleviate training-testing discrepancy. Despite its simplicity, our T2M-GPT shows better performance than competitive approaches, including recent diffusion-based approaches. For example, on HumanML3D, which is currently the largest dataset, we achieve comparable performance on the consistency between text and generated motion (R-Precision), but with FID 0.116 largely outperforming MotionDiffuse of 0.630. Additionally, we conduct analyses on HumanML3D and observe that the dataset size is a limitation of our approach. Our work suggests that VQ-VAE still remains a competitive approach for human motion generation. Our implementation is available on the project page: https://mael-zys.github.io/T2M-GPT/.

Recent years, local differential privacy (LDP) has been adopted by many web service providers like Google [23], Apple [33] and Microsoft [15] to collect and analyse users’ data privately. In this paper, we consider the problem of discrete distribution estimation under local differential privacy constraints. Distribution estimation is one of the most fundamental estimation problems, which is widely studied in both non-private and private settings. In the local model, private mechanisms with provably optimal sample complexity are known. However, they are optimal only in the worst-case sense; their sample complexity is proportional to the size of the entire universe, which could be huge in practice. In this paper, we consider sparse or approximately sparse (e.g. highly skewed) distribution, and show that the number of samples needed could be significantly reduced. This problem has been studied recently [1], but they only consider strict sparse distributions and the high privacy regime. We propose new privatization mechanisms based on compressive sensing. Our methods work for approximately sparse distributions and medium privacy, and have optimal sample and communication complexity.

Camera self-calibration techniques make it possible to compute three-dimensional measurements of an observed scene from video sequences without prior knowledge of the camera intrinsic parameters. However, most previous techniques make the overly constraint assumption that camera intrinsic parameters remain constant throughout the image sequence. The self-calibration method described in this article generalises the camera model by adding into the intrinsic parameter set a variable focal length, whose change is approximated by a cubic B-spline. The modified intrinsic parameters are then calculated with adapted Kruppa equations, together with an initialization procedure that utilizes the genetic algorithm. Results obtained using synthetic image sequences are presented. It shows the algorithm's ability to trade the changing focal length (zooming) even under noisy conditions.