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.

We present EgoTAP, a heatmap-to-3D pose lifting method for highly accurate stereo egocentric 3D pose estimation. Severe self-occlusion and out-of-view limbs in egocentric camera views make accurate pose estimation a challenging problem. To address the challenge, prior methods employ joint heatmaps-probabilistic 2D representations of the body pose, but heatmap-to-3D pose conversion still remains an inaccurate process. We propose a novel heatmap-to-3D lifting method composed of the Grid ViT Encoder and the Propagation Network. The Grid ViT Encoder summarizes joint heatmaps into effective feature embedding using self-attention. Then, the Propagation Network estimates the 3D pose by utilizing skeletal information to better estimate the position of obscure joints. Our method significantly outperforms the previous state-of-the-art qualitatively and quantitatively demonstrated by a 23.9% reduction of error in an MPJPE metric. Our source code is available on GitHub.

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.

Learning deep neural networks on human motion data has become common in computer graphics research, but the heterogeneity of available datasets poses challenges for training large-scale networks. This paper presents a framework that allows us to solve various animation tasks in a skeleton-agnostic manner. The core of our framework is to learn an embedding space to disentangle skeleton-related information from input motion while preserving semantics, which we call Skeleton-Agnostic Motion Embedding (SAME). To efficiently learn the embedding space, we develop a novel autoencoder with graph convolution networks and provide new formulations of various animation tasks operating in the SAME space. We showcase various examples, including retargeting, reconstruction, and interactive character control, and conduct an ablation study to validate design choices made during development.

What happens if we put vision and touch into conflict? Which modality "wins"? Although several previous studies have addressed this topic, they have solely focused on integration of vision and touch for low-level object properties (such as curvature, slant, or depth). In the present study, we introduce a multimodal mixed-reality setup based on real-time hand-tracking, which was used to display real-world, haptic exploration of objects in a virtual environment through a head-mounted-display (HMD). With this setup we studied multimodal conflict situations of objects varying along higher-level, parametrically-controlled global shape properties. Participants explored these objects in both unimodal and multimodal settings with the latter including congruent and incongruent conditions and differing instructions for weighting the input modalities. Results demonstrated a surprisingly clear touch dominance throughout all experiments, which in addition was only marginally influenceable through instructions to bias their modality weighting. We also present an initial analysis of the hand-tracking patterns that illustrates the potential for our setup to investigate exploration behavior in more detail.