3D Gaussian Splatting (3DGS) has enabled photorealistic and real-time rendering of 3D head avatars. Existing 3DGS-based avatars typically rely on tens of thousands of 3D Gaussian points (Gaussians), with the number of Gaussians fixed after training. However, many practical applications require adjustable levels of detail (LOD) to balance rendering efficiency and visual quality. In this work, we propose "ArchitectHead", the first framework for creating 3D Gaussian head avatars that support continuous control over LOD. Our key idea is to parameterize the Gaussians in a 2D UV feature space and propose a UV feature field composed of multi-level learnable feature maps to encode their latent features. A lightweight neural network-based decoder then transforms these latent features into 3D Gaussian attributes for rendering. ArchitectHead controls the number of Gaussians by dynamically resampling feature maps from the UV feature field at the desired resolutions. This method enables efficient and continuous control of LOD without retraining. Experimental results show that ArchitectHead achieves state-of-the-art (SOTA) quality in self and cross-identity reenactment tasks at the highest LOD, while maintaining near SOTA performance at lower LODs. At the lowest LOD, our method uses only 6.2% of the Gaussians while the quality degrades moderately (L1 Loss +7.9%, PSNR −0.97%, SSIM −0.6%, LPIPS Loss +24.1%), and the rendering speed nearly doubles. Project homepage: https://peizhiyan.github.io/docs/architect/.

Existing image-based 3D face reconstruction methods rely on a virtual camera to project the reconstructed 3D face onto the 2D image plane for comparison with the input image, a crucial step for accurate results. To simplify the reconstruction process, these methods often use fixed camera intrinsics and assume minimal perspective distortion, overlooking the varying distortion levels in "in-the-wild" images and leading to inaccuracies in reconstructed 3D face shapes. To address this issue, we propose estimating the virtual camera’s optimal field-of-view (FOV) for a given image, enabling consistent 3D face reconstruction across varying distortion levels. We introduce two synthetic datasets: one to train our FOV estimation network (FOV-Net) and another to evaluate its performance and reconstruction accuracy. We use the FOV-Net predicted FOV to initialize the camera, which is used in the fitting-based reconstruction process. Experiments show that our approach significantly improves reconstruction consistency under different levels of perspective distortion.

Recent advancements in 3D Gaussian Splatting (3DGS) have unlocked significant potential for modeling 3D head avatars, providing greater flexibility than mesh-based methods and more efficient rendering compared to NeRF-based approaches. Despite these advancements, the creation of controllable 3DGS-based head avatars remains time-intensive, often requiring tens of minutes to hours. To expedite this process, we here introduce the “Gaussian Déjà-vu” framework, which first obtains a generalized model of the head avatar and then personalizes the result. The generalized model is trained on large 2D (synthetic and real) image datasets. This model provides a well-initialized 3D Gaussian head that is further refined using a monocular video to achieve the personalized head avatar. For personalizing, we propose learnable expression-aware rectification blendmaps to correct the initial 3D Gaussians, ensuring rapid convergence without the reliance on neural networks. Experiments demonstrate that the proposed method meets its objectives. It outperforms state-of-the-art 3D Gaussian head avatars in terms of photorealistic quality as well as reduces training time consumption to at least a quarter of the existing methods, producing the avatar in minutes. Project homepage: https://peizhiyan.github.io/docs/dejavu

3D Face shape stylization refers to transforming a realistic 3D face shape into a different style, such as a cartoon face style. To solve this problem, this paper proposes modeling this task as a deformation transfer problem. This approach significantly reduces labor costs, as the artists would only need to create a single template for each face style. Realistic facial features of the original 3D face e.g. the nose or chin shape, would thus be automatically transferred to those in the style template. Deformation transfer methods, however, have two drawbacks. They are slow and they require re-optimization for every new input face. To address these weaknesses, we propose a neural network-based 3D face shape stylization method. This method is trained through weakly supervised learning, and its template's structure is preserved using our novel template-guided mesh smoothing regularization. Our method is the first learning-based deformation transfer method for 3D face shape stylization. Its employment offers the useful and practical benefit of not requiring paired training data. The experiments show that the quality of the stylized faces obtained by our method is comparable to that of the traditional deformation transfer method, achieving an average Chamfer Distance of approximately 0.01 mm. However, our approach significantly boosts the processing speed, achieving a rate approximately 3,000 times faster than the traditional deformation transfer.

The 3D-aware parametric face model named HeadNeRF achieved advantages in rendering photo-realistic face images. However, it has two limitations: (1) it uses single-image fitting reconstruction that is slow and prone to overfitting; (2) it lacks explicit 3D geometry information, making using semantic facial-parts-based loss challenging. This paper presents a 3D-aware face reconstruction learning framework tailored for HeadNeRF to address the limitations. We train a face encoder network that can directly learn the disentangled features for facial reconstruction to address the first limitation. For the second limitation, we introduce a lightweight semantic face segmentation network and facial-parts-based loss function to improve the reconstruction accuracy and quality. Our experiments show that the proposed method achieves a low reconstruction time consumption and enhanced reconstruction accuracy. Project page: https://peizhiyan.github.io/docs/headnerf+

In a mobile edge computing (MEC) network, mobile devices could selectively offload tasks to the edge server(s) to save time and energy. However, we should consider many dynamic factors in task offloading optimization, which increases the complexity of this problem. Instead of executing the traditional optimization algorithm repeatedly, a well-trained empirical model such as an artificial neural network could be more efficient in decision making. In this research, considering the potential uneven spatial distribution of mobile devices in an MEC network with multiple wireless edge gateways, we allow an edge gateway to offload tasks to a nearby edge gateway further. We propose a deep reinforcement learning-based joint optimization approach for both device-level and edge-level task offloading. Experimental results show that the proposed approach achieves a near-optimal task delay performance and a better trade-off between the task delay and the energy consumption on tasks.

Radio Frequency Identification (RFID) plays a key role in the Internet of things (IoT). The type of scenario that needs to use many readers to cover a large area is a dense RFID environment scenario. In supply-chain management, companies such as Wal-Mart use dense RFID reader systems to track products [1]. Collisions usually happen in dense RFID reader systems, which reduce the number of tags that can be read by the system. Many algorithms were designed to eliminate the collisions in a dense RFID environment. A Maximum-Weight-Independent-Set-Based Algorithm (MWISBA) [2] is used to solve the dense RFID readers' arrangement uses a graph-based algorithm to get the MWIS. However, MWISBA does not consider interference range, it can only avoid reader-to-tag collisions. Based on MWISBA, an improved algorithm called MWISBAII [3] can avoid both reader-to-tag collisions and reader-to-reader collisions. However, both MWISBA and MWISBAII are centralized algorithms. In this paper, we propose a distributed realization of MWISBAII. In our distributed algorithm, each reader can communicate with other neighbor readers to share and collect information; making the local decision afterwards. The experimental results show that our distributed algorithm can get almost the same performance as the MWISBAII.