Takin: A Cohort of Superior Quality Zero-shot Speech Generation Models

With the advent of the big data and large language model era, zero-shot personalized rapid customization has emerged as a significant trend. In this report, we introduce Takin AudioLLM, a series of techniques and models, mainly including Takin TTS, Takin VC, and Takin Morphing, specifically designed for audiobook production. These models are capable of zero-shot speech production, generating high-quality speech that is nearly indistinguishable from real human speech and facilitating individuals to customize the speech content according to their own needs. Specifically, we first introduce Takin TTS, a neural codec language model that builds upon an enhanced neural speech codec and a multi-task training framework, capable of generating high-fidelity natural speech in a zero-shot way. For Takin VC, we advocate an effective content and timbre joint modeling approach to improve the speaker similarity, while advocating for a conditional flow matching based decoder to further enhance its naturalness and expressiveness. Last, we propose the Takin Morphing system with highly decoupled and advanced timbre and prosody modeling approaches, which enables individuals to customize speech production with their preferred timbre and prosody in a precise and controllable manner. Extensive experiments validate the effectiveness and robustness of our Takin AudioLLM series models. For detailed demos, please refer to https://takinaudiollm.github.io.

ConDense: Consistent 2D/3D Pre-training for Dense and Sparse Features from Multi-View Images

To advance the state of the art in the creation of 3D foundation models, this paper introduces the ConDense framework for 3D pre-training utilizing existing pre-trained 2D networks and large-scale multi-view datasets. We propose a novel 2D-3D joint training scheme to extract co-embedded 2D and 3D features in an end-to-end pipeline, where 2D-3D feature consistency is enforced through a volume rendering NeRF-like ray marching process. Using dense per pixel features we are able to 1) directly distill the learned priors from 2D models to 3D models and create useful 3D backbones, 2) extract more consistent and less noisy 2D features, 3) formulate a consistent embedding space where 2D, 3D, and other modalities of data (e.g., natural language prompts) can be jointly queried. Furthermore, besides dense features, ConDense can be trained to extract sparse features (e.g., key points), also with 2D-3D consistency -- condensing 3D NeRF representations into compact sets of decorated key points. We demonstrate that our pre-trained model provides good initialization for various 3D tasks including 3D classification and segmentation, outperforming other 3D pre-training methods by a significant margin. It also enables, by exploiting our sparse features, additional useful downstream tasks, such as matching 2D images to 3D scenes, detecting duplicate 3D scenes, and querying a repository of 3D scenes through natural language -- all quite efficiently and without any per-scene fine-tuning.

PIE: Parkour with Implicit-Explicit Learning Framework for Legged Robots

Parkour presents a highly challenging task for legged robots, requiring them to traverse various terrains with agile and smooth locomotion. This necessitates comprehensive understanding of both the robot's own state and the surrounding terrain, despite the inherent unreliability of robot perception and actuation. Current state-of-the-art methods either rely on complex pre-trained high-level terrain reconstruction modules or limit the maximum potential of robot parkour to avoid failure due to inaccurate perception. In this paper, we propose a one-stage end-to-end learning-based parkour framework: Parkour with Implicit-Explicit learning framework for legged robots (PIE) that leverages dual-level implicit-explicit estimation. With this mechanism, even a low-cost quadruped robot equipped with an unreliable egocentric depth camera can achieve exceptional performance on challenging parkour terrains using a relatively simple training process and reward function. While the training process is conducted entirely in simulation, our real-world validation demonstrates successful zero-shot deployment of our framework, showcasing superior parkour performance on harsh terrains.

Constrained Behavior Cloning for Robotic Learning

Behavior cloning (BC) is a popular supervised imitation learning method in the societies of robotics, autonomous driving, etc., wherein complex skills can be learned by direct imitation from expert demonstrations. Despite its rapid development, it is still affected by limited field of view where accumulation of sensors and joint noise bring compounding errors. In this paper, we introduced geometrically and historically constrained behavior cloning (GHCBC) to dominantly consider high-level state information inspired by neuroscientists, wherein the geometrically constrained behavior cloning were used to geometrically constrain predicting poses, and the historically constrained behavior cloning were utilized to temporally constrain action sequences. The synergy between these two types of constrains enhanced the BC performance in terms of robustness and stability. Comprehensive experimental results showed that success rates were improved by 29.73% in simulation and 39.4% in real robot experiments in average, respectively, compared to state-of-the-art BC method, especially in long-term operational scenes, indicating great potential of using the GHCBC for robotic learning.

NeuroSEM: A hybrid framework for simulating multiphysics problems by coupling PINNs and spectral elements

Multiphysics problems that are characterized by complex interactions among fluid dynamics, heat transfer, structural mechanics, and electromagnetics, are inherently challenging due to their coupled nature. While experimental data on certain state variables may be available, integrating these data with numerical solvers remains a significant challenge. Physics-informed neural networks (PINNs) have shown promising results in various engineering disciplines, particularly in handling noisy data and solving inverse problems. However, their effectiveness in forecasting nonlinear phenomena in multiphysics regimes is yet to be fully established. This study introduces NeuroSEM, a hybrid framework integrating PINNs with the high-fidelity Spectral Element Method (SEM) solver, Nektar++. NeuroSEM leverages strengths of both PINNs and SEM, providing robust solutions for multiphysics problems. PINNs are trained to assimilate data and model physical phenomena in specific subdomains, which are then integrated into Nektar++. We demonstrate the efficiency and accuracy of NeuroSEM for thermal convection in cavity flow and flow past a cylinder. The framework effectively handles data assimilation by addressing those subdomains and state variables where data are available. We applied NeuroSEM to the Rayleigh-B\'enard convection system, including cases with missing thermal boundary conditions. Our results indicate that NeuroSEM accurately models the physical phenomena and assimilates the data within the specified subdomains. The framework's plug-and-play nature facilitates its extension to other multiphysics or multiscale problems. Furthermore, NeuroSEM is optimized for an efficient execution on emerging integrated GPU-CPU architectures. This hybrid approach enhances the accuracy and efficiency of simulations, making it a powerful tool for tackling complex engineering challenges in various scientific domains.

Pseudo-Labeling by Multi-Policy Viewfinder Network for Image Cropping

Automatic image cropping models predict reframing boxes to enhance image aesthetics. Yet, the scarcity of labeled data hinders the progress of this task. To overcome this limitation, we explore the possibility of utilizing both labeled and unlabeled data together to expand the scale of training data for image cropping models. This idea can be implemented in a pseudo-labeling way: producing pseudo labels for unlabeled data by a teacher model and training a student model with these pseudo labels. However, the student may learn from teacher's mistakes. To address this issue, we propose the multi-policy viewfinder network (MPV-Net) that offers diverse refining policies to rectify the mistakes in original pseudo labels from the teacher. The most reliable policy is selected to generate trusted pseudo labels. The reliability of policies is evaluated via the robustness against box jittering. The efficacy of our method can be evaluated by the improvement compared to the supervised baseline which only uses labeled data. Notably, our MPV-Net outperforms off-the-shelf pseudo-labeling methods, yielding the most substantial improvement over the supervised baseline. Furthermore, our approach achieves state-of-the-art results on both the FCDB and FLMS datasets, signifying the superiority of our approach.

A comprehensive and FAIR comparison between MLP and KAN representations for differential equations and operator networks

Kolmogorov-Arnold Networks (KANs) were recently introduced as an alternative representation model to MLP. Herein, we employ KANs to construct physics-informed machine learning models (PIKANs) and deep operator models (DeepOKANs) for solving differential equations for forward and inverse problems. In particular, we compare them with physics-informed neural networks (PINNs) and deep operator networks (DeepONets), which are based on the standard MLP representation. We find that although the original KANs based on the B-splines parameterization lack accuracy and efficiency, modified versions based on low-order orthogonal polynomials have comparable performance to PINNs and DeepONet although they still lack robustness as they may diverge for different random seeds or higher order orthogonal polynomials. We visualize their corresponding loss landscapes and analyze their learning dynamics using information bottleneck theory. Our study follows the FAIR principles so that other researchers can use our benchmarks to further advance this emerging topic.

Low-Cost and Real-Time Industrial Human Action Recognitions Based on Large-Scale Foundation Models

Industrial managements, including quality control, cost and safety optimization, etc., heavily rely on high quality industrial human action recognitions (IHARs) which were hard to be implemented in large-scale industrial scenes due to their high costs and poor real-time performance. In this paper, we proposed a large-scale foundation model(LSFM)-based IHAR method, wherein various LSFMs and lightweight methods were jointly used, for the first time, to fulfill low-cost dataset establishment and real-time IHARs. Comprehensive tests on in-situ large-scale industrial manufacturing lines elucidated that the proposed method realized great reduction on employment costs, superior real-time performance, and satisfactory accuracy and generalization capabilities, indicating its great potential as a backbone IHAR method, especially for large-scale industrial applications.

Toward Understanding Key Estimation in Learning Robust Humanoid Locomotion

Accurate state estimation plays a critical role in ensuring the robust control of humanoid robots, particularly in the context of learning-based control policies for legged robots. However, there is a notable gap in analytical research concerning estimations. Therefore, we endeavor to further understand how various types of estimations influence the decision-making processes of policies. In this paper, we provide quantitative insight into the effectiveness of learned state estimations, employing saliency analysis to identify key estimation variables and optimize their combination for humanoid locomotion tasks. Evaluations assessing tracking precision and robustness are conducted on comparative groups of policies with varying estimation combinations in both simulated and real-world environments. Results validated that the proposed policy is capable of crossing the sim-to-real gap and demonstrating superior performance relative to alternative policy configurations.

Digging Into Normal Incorporated Stereo Matching

Despite the remarkable progress facilitated by learning-based stereo-matching algorithms, disparity estimation in low-texture, occluded, and bordered regions still remains a bottleneck that limits the performance. To tackle these challenges, geometric guidance like plane information is necessary as it provides intuitive guidance about disparity consistency and affinity similarity. In this paper, we propose a normal incorporated joint learning framework consisting of two specific modules named non-local disparity propagation(NDP) and affinity-aware residual learning(ARL). The estimated normal map is first utilized for calculating a non-local affinity matrix and a non-local offset to perform spatial propagation at the disparity level. To enhance geometric consistency, especially in low-texture regions, the estimated normal map is then leveraged to calculate a local affinity matrix, providing the residual learning with information about where the correction should refer and thus improving the residual learning efficiency. Extensive experiments on several public datasets including Scene Flow, KITTI 2015, and Middlebury 2014 validate the effectiveness of our proposed method. By the time we finished this work, our approach ranked 1st for stereo matching across foreground pixels on the KITTI 2015 dataset and 3rd on the Scene Flow dataset among all the published works.