3D Gaussian Splatting (3DGS) has attracted significant attention for its high-quality novel view rendering, inspiring research to address real-world challenges. While conventional methods depend on sharp images for accurate scene reconstruction, real-world scenarios are often affected by defocus blur due to finite depth of field, making it essential to account for realistic 3D scene representation. In this study, we propose CoCoGaussian, a Circle of Confusion-aware Gaussian Splatting that enables precise 3D scene representation using only defocused images. CoCoGaussian addresses the challenge of defocus blur by modeling the Circle of Confusion (CoC) through a physically grounded approach based on the principles of photographic defocus. Exploiting 3D Gaussians, we compute the CoC diameter from depth and learnable aperture information, generating multiple Gaussians to precisely capture the CoC shape. Furthermore, we introduce a learnable scaling factor to enhance robustness and provide more flexibility in handling unreliable depth in scenes with reflective or refractive surfaces. Experiments on both synthetic and real-world datasets demonstrate that CoCoGaussian achieves state-of-the-art performance across multiple benchmarks.

Recent works on dynamic 3D neural field reconstruction assume the input from synchronized multi-view videos whose poses are known. The input constraints are often not satisfied in real-world setups, making the approach impractical. We show that unsynchronized videos from unknown poses can generate dynamic neural fields as long as the videos capture human motion. Humans are one of the most common dynamic subjects captured in videos, and their shapes and poses can be estimated using state-of-the-art libraries. While noisy, the estimated human shape and pose parameters provide a decent initialization point to start the highly non-convex and under-constrained problem of training a consistent dynamic neural representation. Given the shape and pose parameters of humans in individual frames, we formulate methods to calculate the time offsets between videos, followed by camera pose estimations that analyze the 3D joint positions. Then, we train the dynamic neural fields employing multiresolution grids while we concurrently refine both time offsets and camera poses. The setup still involves optimizing many parameters; therefore, we introduce a robust progressive learning strategy to stabilize the process. Experiments show that our approach achieves accurate spatio-temporal calibration and high-quality scene reconstruction in challenging conditions.

Video large language models (LLMs) achieve strong video understanding by leveraging a large number of spatio-temporal tokens, but suffer from quadratic computational scaling with token count. To address this, we propose a training-free spatio-temporal token merging method, named STTM. Our key insight is to exploit local spatial and temporal redundancy in video data which has been overlooked in prior work. STTM first transforms each frame into multi-granular spatial tokens using a coarse-to-fine search over a quadtree structure, then performs directed pairwise merging across the temporal dimension. This decomposed merging approach outperforms existing token reduction methods across six video QA benchmarks. Notably, STTM achieves a 2$\times$ speed-up with only a 0.5% accuracy drop under a 50% token budget, and a 3$\times$ speed-up with just a 2% drop under a 30% budget. Moreover, STTM is query-agnostic, allowing KV cache reuse across different questions for the same video. The project page is available at https://www.jshyun.me/projects/sttm.

3D Gaussian Splatting (3DGS) has gained significant attention due to its high-quality novel view rendering, motivating research to address real-world challenges. A critical issue is the camera motion blur caused by movement during exposure, which hinders accurate 3D scene reconstruction. In this study, we propose CoMoGaussian, a Continuous Motion-Aware Gaussian Splatting that reconstructs precise 3D scenes from motion-blurred images while maintaining real-time rendering speed. Considering the complex motion patterns inherent in real-world camera movements, we predict continuous camera trajectories using neural ordinary differential equations (ODEs). To ensure accurate modeling, we employ rigid body transformations, preserving the shape and size of the object but rely on the discrete integration of sampled frames. To better approximate the continuous nature of motion blur, we introduce a continuous motion refinement (CMR) transformation that refines rigid transformations by incorporating additional learnable parameters. By revisiting fundamental camera theory and leveraging advanced neural ODE techniques, we achieve precise modeling of continuous camera trajectories, leading to improved reconstruction accuracy. Extensive experiments demonstrate state-of-the-art performance both quantitatively and qualitatively on benchmark datasets, which include a wide range of motion blur scenarios, from moderate to extreme blur.

In a retrieval system, simultaneously achieving search accuracy and efficiency is inherently challenging. This challenge is particularly pronounced in partially relevant video retrieval (PRVR), where incorporating more diverse context representations at varying temporal scales for each video enhances accuracy but increases computational and memory costs. To address this dichotomy, we propose a prototypical PRVR framework that encodes diverse contexts within a video into a fixed number of prototypes. We then introduce several strategies to enhance text association and video understanding within the prototypes, along with an orthogonal objective to ensure that the prototypes capture a diverse range of content. To keep the prototypes searchable via text queries while accurately encoding video contexts, we implement cross- and uni-modal reconstruction tasks. The cross-modal reconstruction task aligns the prototypes with textual features within a shared space, while the uni-modal reconstruction task preserves all video contexts during encoding. Additionally, we employ a video mixing technique to provide weak guidance to further align prototypes and associated textual representations. Extensive evaluations on TVR, ActivityNet-Captions, and QVHighlights validate the effectiveness of our approach without sacrificing efficiency.

Most of the existing Video Anomaly Detection (VAD) studies have been conducted within single-domain learning, where training and evaluation are performed on a single dataset. However, the criteria for abnormal events differ across VAD datasets, making it problematic to apply a single-domain model to other domains. In this paper, we propose a new task called Multi-Domain learning forVAD (MDVAD) to explore various real-world abnormal events using multiple datasets for a general model. MDVAD involves training on datasets from multiple domains simultaneously, and we experimentally observe that Abnormal Conflicts between domains hinder learning and generalization. The task aims to address two key objectives: (i) better distinguishing between general normal and abnormal events across multiple domains, and (ii) being aware of ambiguous abnormal conflicts. This paper is the first to tackle abnormal conflict issue and introduces a new benchmark, baselines, and evaluation protocols for MDVAD. As baselines, we propose a framework with Null(Angular)-Multiple Instance Learning and an Abnormal Conflict classifier. Through experiments on a MDVAD benchmark composed of six VAD datasets and using four different evaluation protocols, we reveal abnormal conflicts and demonstrate that the proposed baseline effectively handles these conflicts, showing robustness and adaptability across multiple domains.

Recent self-supervised video representation learning methods focus on maximizing the similarity between multiple augmented views from the same video and largely rely on the quality of generated views. However, most existing methods lack a mechanism to prevent representation learning from bias towards static information in the video. In this paper, we propose frequency augmentation (FreqAug), a spatio-temporal data augmentation method in the frequency domain for video representation learning. FreqAug stochastically removes specific frequency components from the video so that learned representation captures essential features more from the remaining information for various downstream tasks. Specifically, FreqAug pushes the model to focus more on dynamic features rather than static features in the video via dropping spatial or temporal low-frequency components. To verify the generality of the proposed method, we experiment with FreqAug on multiple self-supervised learning frameworks along with standard augmentations. Transferring the improved representation to five video action recognition and two temporal action localization downstream tasks shows consistent improvements over baselines.

Temporal action detection aims to predict the time intervals and the classes of action instances in the video. Despite the promising performance, existing two-stream models exhibit slow inference speed due to their reliance on computationally expensive optical flow. In this paper, we introduce a decomposed cross-modal distillation framework to build a strong RGB-based detector by transferring knowledge of the motion modality. Specifically, instead of direct distillation, we propose to separately learn RGB and motion representations, which are in turn combined to perform action localization. The dual-branch design and the asymmetric training objectives enable effective motion knowledge transfer while preserving RGB information intact. In addition, we introduce a local attentive fusion to better exploit the multimodal complementarity. It is designed to preserve the local discriminability of the features that is important for action localization. Extensive experiments on the benchmarks verify the effectiveness of the proposed method in enhancing RGB-based action detectors. Notably, our framework is agnostic to backbones and detection heads, bringing consistent gains across different model combinations.

This paper addresses the problem of three-dimensional (3D) human mesh estimation in complex poses and occluded situations. Although many improvements have been made in 3D human mesh estimation using the two-dimensional (2D) pose with occlusion between humans, occlusion from complex poses and other objects remains a consistent problem. Therefore, we propose the novel Skinned Multi-Person Linear (SMPL) Edge Feature Distillation (SEFD) that demonstrates robustness to complex poses and occlusions, without increasing the number of parameters compared to the baseline model. The model generates an SMPL overlapping edge similar to the ground truth that contains target person boundary and occlusion information, performing subsequent feature distillation in a simple edge map. We also perform experiments on various benchmarks and exhibit fidelity both qualitatively and quantitatively. Extensive experiments prove that our method outperforms the state-of-the-art method by 2.8% in MPJPE and 1.9% in MPVPE on a benchmark 3DPW dataset in the presence of domain gap. Also, our method is superior in 3DPW-OCC, 3DPW-PC, RH-Dataset, OCHuman, Crowd-Pose, and LSP dataset in which occlusion, complex pose, and domain gap exist. The code and occlusion & complex pose annotation will be available at https://github.com/YangChangHee/ICCV2023_SEFD_RELEASE/.

Semi-supervised video object segmentation (VOS) methods aim to segment target objects with the help of pixel-level annotations in the first frame. Many methods employ Transformer-based attention modules to propagate the given annotations in the first frame to the most similar patch or pixel in the following frames. Although they have shown impressive results, they can still be prone to errors in challenging scenes with multiple overlapping objects. To tackle this problem, we propose an object-centric VOS (OCVOS) method that exploits query-based Transformer decoder blocks. After aggregating target object information with typical matching-based approaches, the Transformer networks extract object-wise information by interacting with object queries. In this way, the proposed method considers not only global and contextual information but also object-centric representations. We validate its effectiveness in inducing object-wise information compared to existing methods on the DAVIS and YouTube-VOS benchmarks.

In existing joint detection and tracking methods, pairwise relational features are used to match previous track-lets to current detections. However, the features may not be discriminative enough for a tracker to identify a target from a large number of detections. Selecting only high-scored detections for tracking may lead to missed detections whose confidence score is low. Consequently, in the on-line setting, this results in disconnections of tracklets which cannot be recovered. In this regard, we present Sparse Graph Tracker (SGT), a novel online graph tracker using higher-order relational features which are more discriminative by aggregating the features of neighboring detections and their relations. SGT converts video data into a graph where detections, their connections, and the relational features of two connected nodes are represented by nodes, edges, and edge features, respectively. The strong edge features allow SGT to track targets with tracking candidates selected by top-K scored detections with large K. As a result, even low-scored detections can be tracked, and the missed detections are also recovered. The robustness of K value is shown through the extensive experiments. In the MOT16/17/20 and HiEve Challenge, SGT outperforms the state-of-the-art trackers with real-time inference speed. Especially, a large improvement in MOTA is shown in the MOT20 and HiEve Challenge. Code is available at https://github.com/HYUNJS/SGT.

Deep neural networks for video action recognition frequently require 3D convolutional filters and often encounter overfitting due to a larger number of parameters. In this paper, we propose Random Mean Scaling (RMS), a simple and effective regularization method, to relieve the overfitting problem in 3D residual networks. The key idea of RMS is to randomly vary the magnitude of low-frequency components of the feature to regularize the model. The low-frequency component can be derived by a spatio-temporal mean on the local patch of a feature. We present that selective regularization on this locally smoothed feature makes a model handle the low-frequency and high-frequency component distinctively, resulting in performance improvement. RMS can enhance a model with little additional computation only during training, similar to other regularization methods. RMS also can be incorporated into typical training process without any bells and whistles. Experimental results show the improvement in generalization performance on a popular action recognition datasets demonstrating the effectiveness of RMS as a regularization technique, compared to other state-of-the-art regularization methods.

Most existing multi-person tracking approaches rely on appearance based re-identification (re-ID) to resolve fragmented tracklets. However, simply using appearance information could be insufficient for videos containing severe pose changes, such as sports or dance videos. With the goal of learning pose-invariant representations, we propose an end-to-end deep learning framework Sparse-Temporal ReID Network. Our proposed network not only realizes human pose disentanglement in an image recovery manner, but also makes efficient linkages between the identical subjects via a unique Sparse temporal identity sampling technique across time steps. Experimental results demonstrate the effectiveness of our proposed method on both multi-view re-ID benchmarks and our newly collected dance video dataset DanceReID1.