Existing 4D Gaussian Splatting (4DGS) methods struggle to accurately reconstruct dynamic scenes, often failing to resolve ambiguous pixel correspondences and inadequate densification in dynamic regions. We address these issues by introducing a novel method composed of two key components: (1) Elliptical Error Clustering and Error Correcting Splat Addition that pinpoints dynamic areas to improve and initialize fitting splats, and (2) Grouped 4D Gaussian Splatting that improves consistency of mapping between splats and represented dynamic objects. Specifically, we classify rendering errors into missing-color and occlusion types, then apply targeted corrections via backprojection or foreground splitting guided by cross-view color consistency. Evaluations on Neural 3D Video and Technicolor datasets demonstrate that our approach significantly improves temporal consistency and achieves state-of-the-art perceptual rendering quality, improving 0.39dB of PSNR on the Technicolor Light Field dataset. Our visualization shows improved alignment between splats and dynamic objects, and the error correction method’s capability to identify errors and properly initialize new splats. Our implementation details and source code are available at https://github.com/tho-kn/cem-4dgs.

Object detection, while being an attractive interaction method for Augmented Reality (AR), is fundamentally error-prone due to the probabilistic nature of the underlying AI models, resulting in sub-optimal user experiences. In this paper, we explore the effect of three game design concepts, Ambiguity, Transparency, and Controllability, to provide better gameplay experiences in AR games that use error-prone object detection-based interaction modalities. First, we developed a base AR pet breeding game, called Bubbleu that uses object detection as a key interaction method. We then implemented three different variants, each according to the three concepts, to investigate the impact of each design concept on the overall user experience. Our user study results show that each design has its own strengths and can improve player experiences in different ways such as decreasing perceived errors (Ambiguity), explaining the system (Transparency), and enabling users to control the rate of uncertainties (Controllability).

We present FarfetchFusion, a fully mobile live 3D telepresence system. Enabling mobile live telepresence is a challenging problem as it requires i) realistic reconstruction of the user and ii) high responsiveness for immersive experience. We first thoroughly analyze the live 3D telepresence pipeline and identify three critical challenges: i) 3D data streaming latency and compression complexity, ii) computational complexity of volumetric fusion-based 3D reconstruction, and iii) inconsistent reconstruction quality due to sparsity of mobile 3D sensors. To tackle the challenges, we propose a disentangled fusion approach, which separates invariant regions and dynamically changing regions with our low-complexity spatio-temporal alignment technique, topology anchoring. We then design and implement an end-to-end system, which achieves realistic reconstruction quality comparable to existing server-based solutions while meeting the real-time performance requirements (<100 ms end-to-end latency, 30 fps throughput, <16 ms motion-to-photon latency) solely relying on mobile computation capability.

We present Ego3DPose, a highly accurate binocular egocentric 3D pose reconstruction system. The binocular egocentric setup offers practicality and usefulness in various applications, however, it remains largely under-explored. It has been suffering from low pose estimation accuracy due to viewing distortion, severe self-occlusion, and limited field-of-view of the joints in egocentric 2D images. Here, we notice that two important 3D cues, stereo correspondences, and perspective, contained in the egocentric binocular input are neglected. Current methods heavily rely on 2D image features, implicitly learning 3D information, which introduces biases towards commonly observed motions and leads to low overall accuracy. We observe that they not only fail in challenging occlusion cases but also in estimating visible joint positions. To address these challenges, we propose two novel approaches. First, we design a two-path network architecture with a path that estimates pose per limb independently with its binocular heatmaps. Without full-body information provided, it alleviates bias toward trained full-body distribution. Second, we leverage the egocentric view of body limbs, which exhibits strong perspective variance (e.g., a significantly large-size hand when it is close to the camera). We propose a new perspective-aware representation using trigonometry, enabling the network to estimate the 3D orientation of limbs. Finally, we develop an end-to-end pose reconstruction network that synergizes both techniques. Our comprehensive evaluations demonstrate that Ego3DPose outperforms state-of-the-art models by a pose estimation error (i.e., MPJPE) reduction of 23.1% in the UnrealEgo dataset. Our qualitative results highlight the superiority of our approach across a range of scenarios and challenges.

We present FlexPatch, a novel mobile system to enable accurate and real-time object detection over high-resolution video streams. A widely-used approach for real-time video analysis is detection-based tracking (DBT), i.e., running the heavy-but-accurate detector every few frames and applying a lightweight tracker for in-between frames. However, the approach is limited for real-time processing of high-resolution videos in that i) a lightweight tracker fails to handle occlusion, object appearance changes, and occurrences of new objects, and ii) the detection results do not effectively offset tracking errors due to the high detection latency. We propose tracking-aware patching technique to address such limitations of the DBT frameworks. It effectively identifies a set of subareas where the tracker likely fails and tightly packs them into a small-sized rectangular area where the detection can be efficiently performed at low latency. This prevents the accumulation of tracking errors and offsets the tracking errors with frequent fresh detection results. Our extensive evaluation shows that FlexPatch not only enables real-time and power-efficient analysis of high-resolution frames on mobile devices but also improves the overall accuracy by 146% compared to baseline DBT frameworks.

We present GROOT, a mobile volumetric video streaming system that delivers three-dimensional data to mobile devices for a fully immersive virtual and augmented reality experience. The system design for streaming volumetric videos should be fundamentally different from conventional 2D video streaming systems. First, the amount of data required to deliver the 3D volume is considerably larger than conventional videos with frames of 2D images, even compared to high-resolution 2D or 360° videos. Second, the 3D data representation, which encodes the surface of objects within the volume, is a sparse and unorganized data structure with varying scales, whereas a conventional video is composed of a sequence of images with the fixed-size 2D grid structure. GROOT is a streaming framework with a novel data structure that enables not only real-time transmission and decoding on mobile devices but also continuous on-demand user view adaptation. Specifically, we modify the conventional octree to introduce the independence of leaf nodes with minimal memory overhead, which enables parallel decoding of highly irregular 3D data. We also developed a suite of techniques to compress color information and filter out 3D points outside of a user's view, which efficiently minimizes the data size and decoding cost. Our extensive evaluation shows that GROOT achieves more stable but faster frame rates compared to any previous method to stream and visualize volumetric videos on mobile devices.