Seunggwan Lee

MinHyuk Jang

Sieun Kim

SeungHyun Lee

Wonjeong Ryoo

DongYoon Seo

Taebeom Lee

The lab challenges unsolved real-life problems by inventing novel artificial intelligence for computer vision. To do so, our research focuses on developing innovative computer vision systems through deep learning algorithms and heterogeneous visual sensors. We also adopt machine learning to innovate visual machine intelligence.
In the narrow scope of research, the lab is interested in the following:
– Self-Motivated AI
– Physical AI
– AI Safety
– Foundation Model
– 3D Computer Vision
– Autonomous Driving

Illegal 3D content distribution tracking system based on DNN forensic watermarking

Vision-Guided Targeted Grasping and Vibration for Robotic Pollination in Controlled Environments

Functional Hand Type Prior for 3D Hand Pose Estimation and Action Recognition from Egocentric View Monocular Videos.

Ora3d: Overlap region aware multi-view 3d object detection

Audio-Guided Image Manipulation for Artistic Paintings

Audio-guided implicit neural representation for local image stylization

Dual policy learning for aggregation optimization in graph neural network-based recommender systems

Sound-guided semantic image manipulation

Egocentric Hand Activity Video Dataset and Bidirectional Motion-Priors for Hand Action Recognition

Supplemental Materials to The Power of Sound (TPoS): Audio Reactive Video Generation with Stable Diffusion

3d occupancy prediction with low-resolution queries via prototype-aware view transformation

Test-time adaptation for online vision-language navigation with feedback-based reinforcement learning

Pasta: Part-aware sketch-to-3d shape generation with text-aligned prior

The power of sound (tpos): Audio reactive video generation with stable diffusion

Cmda: Cross-modal and domain adversarial adaptation for lidar-based 3d object detection

Higher-order relational reasoning for pedestrian trajectory prediction

Insightful instance features for 3D instance segmentation

Unified domain generalization and adaptation for multi-view 3d object detection

RevoNAD: Reflective Evolutionary Exploration for Neural Architecture Design

MELA: Multi-Event Localization Answering Framework for Video Question Answering

Occlusion-Aware Temporally Consistent Amodal Completion for 3D Human-Object Interaction Reconstruction

Enet: A deep neural network architecture for real-time semantic segmentation

A large-scale annotated mechanical components benchmark for classification and retrieval tasks with deep neural networks

Sound-guided semantic video generation

Learning hand articulations by hallucinating heat distribution

Parrot: Pareto-optimal Multi-Reward Reinforcement Learning Framework for Text-to-Image Generation

MEVG : Multi-event Video Generation with Text-to-Video Models

WateRF: Robust Watermarks in Radiance Fields for Protection of Copyrights

Higher-order Relational Reasoning for Pedestrian Trajectory Prediction

HapMotion: motion-to-tactile framework with wearable haptic devices for immersive VR performance experience

3D-GSW: 3D Gaussian Splatting for Robust Watermarking

EditSplat: Multi-View Fusion and Attention-Guided Optimization for View-Consistent 3D Scene Editing with 3D Gaussian Splatting

VisionTrap: Vision-Augmented Trajectory Prediction Guided by Textual Descriptions

Self-supervised multimodal graph convolutional network for collaborative filtering

Soundini: Sound-guided diffusion for natural video editing

Object synthesis by learning part geometry with surface and volumetric representations

Edge-Aware 3D Instance Segmentation Network with Intelligent Semantic Prior

CMDA: Cross-Modal and Domain Adversarial Adaptation for LiDAR-based 3D Object Detection

Hyung-gun Chi

Eugenio Culurciello

Karthik Ramani

Adam Paszke

Qixing Huang

Nick Winovich

Guang Lin

Sangpil Kim

The ability to perform pixel-wise semantic segmentation in real-time is of paramount importance in mobile applications. Recent deep neural networks aimed at this task have the disadvantage of requiring a large number of floating point operations and have long run-times that hinder their usability. In this paper, we propose a novel deep neural network architecture named ENet (efficient neural network), created specifically for tasks requiring low latency operation. ENet is up to 18 faster, requires 75 less FLOPs, has 79 less parameters, and provides similar or better accuracy to existing models. We have tested it on CamVid, Cityscapes and SUN datasets and report on comparisons with existing state-of-the-art methods, and the trade-offs between accuracy and processing time of a network. We present performance measurements of the proposed architecture on embedded systems and suggest possible software improvements that could make ENet even faster.

We introduce a large-scale annotated mechanical components benchmark for classification and retrieval tasks named Mechanical Components Benchmark (MCB): a large-scale dataset of 3D objects of mechanical components. The dataset enables data-driven feature learning for mechanical components. Exploring the shape descriptor for mechanical components is essential to computer vision and manufacturing applications. However, not much attention has been given on creating annotated mechanical components datasets on a large scale. This is because acquiring 3D models is challenging and annotating mechanical components requires engineering knowledge. Our main contributions are the creation of a large-scale annotated mechanical component benchmark, defining hierarchy taxonomy of mechanical components, and benchmarking the effectiveness of deep learning shape classifiers on the …

The recent success of the generative model shows that leveraging the multi-modal embedding space can manipulate an image using text information. However, manipulating an image with other sources rather than text, such as sound, is not easy due to the dynamic characteristics of the sources. Especially, sound can convey vivid emotions and dynamic expressions of the real world. Here, we propose a framework that directly encodes sound into the multi-modal (image-text) embedding space and manipulates an image from the space. Our audio encoder is trained to produce a latent representation from an audio input, which is forced to be aligned with image and text representations in the multi-modal embedding space. We use a direct latent optimization method based on aligned embeddings for sound-guided image manipulation. We also show that our method can mix different modalities, ie, text and audio, which enrich the variety of the image modification. The experiments on zero-shot audio classification and semantic-level image classification show that our proposed model outperforms other text and sound-guided state-of-the-art methods.

The recent success in StyleGAN demonstrates that pre-trained StyleGAN latent space is useful for realistic video generation. However, the generated motion in the video is usually not semantically meaningful due to the difficulty of determining the direction and magnitude in the StyleGAN latent space. In this paper, we propose a framework to generate realistic videos by leveraging multimodal (sound-image-text) embedding space. As sound provides the temporal contexts of the scene, our framework learns to generate a video that is semantically consistent with sound. First, our sound inversion module maps the audio directly into the StyleGAN latent space. We then incorporate the CLIP-based multimodal embedding space to further provide the audio-visual relationships. Finally, the proposed frame generator learns to find the trajectory in the latent space which is coherent with the corresponding sound and …

In recent years, video generation has become a prominent generative tool and has drawn significant attention. However, there is little consideration in audio-to-video generation, though audio contains unique qualities like temporal semantics and magnitude. Hence, we propose The Power of Sound (TPoS) model to incorporate audio input that includes both changeable temporal semantics and magnitude. To generate video frames, TPoS utilizes a latent stable diffusion model with textual semantic information, which is then guided by the sequential audio embedding from our pretrained Audio Encoder. As a result, this method produces audio reactive video contents. We demonstrate the effectiveness of TPoS across various tasks and compare its results with current state-of-the-art techniques in the field of audio-to-video generation. More examples are available at https://ku-vai. github. io/TPoS/

We propose a robust hand pose estimation method by learning hand articulations from depth features and auxiliary modality features. As an additional modality to depth data, we present a function of geometric properties on the surface of the hand described by heat diffusion. The proposed heat distribution descriptor is robust to identify the keypoints on the surface as it incorporates both the local geometry of the hand and global structural representation at multiple time scales. Along this line, we train our heat distribution network to learn the geometrically descriptive representations from the proposed descriptors with the fingertip position labels. Then the hallucination network is guided to mimic the intermediate responses of the heat distribution modality from a paired depth image. We use the resulting geometrically informed responses together with the discriminative depth features estimated from the depth network to regularize the angle parameters in the refinement network. To this end, we conduct extensive evaluations to validate that the proposed framework is powerful as it achieves state-of-the-art performance.

Recent works have demonstrated that using reinforcement learning (RL) with multiple quality rewards can improve the quality of generated images in text-to-image (T2I) generation. However, manually adjusting reward weights poses challenges and may cause over-optimization in certain metrics. To solve this, we propose Parrot, which addresses the issue through multi-objective optimization and introduces an effective multi-reward optimization strategy to approximate Pareto optimal. Utilizing batch-wise Pareto optimal selection, Parrot automatically identifies the optimal trade-off among different rewards. We use the novel multi-reward optimization algorithm to jointly optimize the T2I model and a prompt expansion network, resulting in significant improvement of image quality and also allow to control the trade-off of different rewards using a reward related prompt during inference. Furthermore, we introduce original …

The advances in the Neural Radiance Fields (NeRF) research offer extensive applications in diverse domains but protecting their copyrights has not yet been researched in depth. Recently NeRF watermarking has been considered one of the pivotal solutions for safely deploying NeRF-based 3D representations. However existing methods are designed to apply only to implicit or explicit NeRF representations. In this work we introduce an innovative watermarking method that can be employed in both representations of NeRF. This is achieved by fine-tuning NeRF to embed binary messages in the rendering process. In detail we propose utilizing the discrete wavelet transform in the NeRF space for watermarking. Furthermore we adopt a deferred back-propagation technique and introduce a combination with the patch-wise loss to improve rendering quality and bit accuracy with minimum trade-offs. We evaluate our method in three different aspects: capacity invisibility and robustness of the embedded watermarks in the 2D-rendered images. Our method achieves state-of-the-art performance with faster training speed over the compared state-of-the-art methods. Project page: https://kuai-lab. github. io/cvpr2024waterf/

Social relations have substantial impacts on the potential trajectories of each individual. Modeling these dynamics has been a central solution for more precise and accurate trajectory forecasting. However previous works ignore the importance ofsocial depth'meaning the influences flowing from different degrees of social relations. In this work we propose HighGraph a graph-based pedestrian relational reasoning method that captures the higher-order dynamics of social interactions. First we construct a collision-aware relation graph based on the agents' observed trajectories. Upon this graph structure we build our core module that aggregates the agent features from diverse social distances. As a result the network is able to model complex social relations thereby yielding more accurate and socially acceptable trajectories. Our HighGraph is a plug-and-play module that can be easily applied to any current trajectory predictors. Extensive experiments with ETH/UCY and SDD datasets demonstrate that our HighGraph noticeably improves the previous state-of-the-art baselines both quantitatively and qualitatively.

We present a novel haptic rendering framework that translates the performer’s motions into wearable vibrotactile feedback for an immersive virtual reality (VR) performance experience. Here, we employ a rendering pipeline that extracts meaningful vibrotactile parameters including intensity and location. We compute these parameters from the performer’s upper-body movements which play a significant role in a dance performance. Therefore, we customize a haptic vest and sleeves to support vibrotactile feedback on the frontal and back parts of the torso and shoulders as well. To capture essential movements from the VR performance, we propose a method called motion salient triangle (MST). MST utilizes key skeleton joints’ movements to compute the associated haptic parameters. Our method supports translating both choreographic and communicative motions into vibrotactile feedback. Through a series of user …

As 3D Gaussian Splatting (3D-GS) gains significant attention and its commercial usage increases, the need for watermarking technologies to prevent unauthorized use of the 3D-GS models and rendered images has become increasingly important. In this paper, we introduce a robust watermarking method for 3D-GS that secures copyright of both the model and its rendered images. Our proposed method remains robust against distortions in rendered images and model attacks while maintaining high rendering quality. To achieve these objectives, we present Frequency-Guided Densification (FGD), which removes 3D Gaussians based on their contribution to rendering quality, enhancing real-time rendering and the robustness of the message. FGD utilizes Discrete Fourier Transform to split 3D Gaussians in high-frequency areas, improving rendering quality. Furthermore, we employ a gradient mask for 3D Gaussians and design a wavelet-subband loss to enhance rendering quality. Our experiments show that our method embeds the message in the rendered images invisibly and robustly against various attacks, including model distortion. Our method achieves superior performance in both rendering quality and watermark robustness while improving real-time rendering efficiency. Project page: https://kuai-lab.github.io/cvpr20253dgsw/

Recent advancements in 3D editing have highlighted the potential of text-driven methods in real-time, user-friendly AR/VR applications. However, current methods rely on 2D diffusion models without adequately considering multi-view information, resulting in multi-view inconsistency. While 3D Gaussian Splatting (3DGS) significantly improves rendering quality and speed, its 3D editing process encounters difficulties with inefficient optimization, as pre-trained Gaussians retain excessive source information, hindering optimization. To address these limitations, we propose EditSplat, a novel 3D editing framework that integrates Multi-view Fusion Guidance (MFG) and Attention-Guided Trimming (AGT). Our MFG ensures multi-view consistency by incorporating essential multi-view information into the diffusion process, leveraging classifier-free guidance from the text-to-image diffusion model and the geometric properties of 3DGS. Additionally, our AGT leverages the explicit representation of 3DGS to selectively prune and optimize 3D Gaussians, enhancing optimization efficiency and enabling precise, semantically rich local edits. Through extensive qualitative and quantitative evaluations, EditSplat achieves superior multi-view consistency and editing quality over existing methods, significantly enhancing overall efficiency.

Predicting future trajectories for other road agents is an essential task for autonomous vehicles. Established trajectory prediction methods primarily use agent tracks generated by a detection and tracking system and HD map as inputs. In this work, we propose a novel method that also incorporates visual input from surround-view cameras, allowing the model to utilize visual cues such as human gazes and gestures, road conditions, vehicle turn signals, etc., which are typically hidden from the model in prior methods. Furthermore, we use textual descriptions generated by a Vision-Language Model (VLM) and refined by a Large Language Model (LLM) as supervision during training to guide the model on what to learn from the input data. Despite using these extra inputs, our method achieves a latency of 53 ms, making it feasible for real-time processing, which is significantly faster than that of previous single-agent …

Collaborative filtering (CF) is a central solution for capturing various user-item relationships in building recommender systems. However, when the relationships are sparsely observed, it is challenging to obtain enough signals to infer precise user preferences. Recent studies have attempted to address the sparsity issue by incorporating multimodal information (e.g., image and text) into CF models. However, existing methods mainly focus on capturing modal-specific user preference with multiple unimodal graphs, ignoring the complex nature of user behavior, which is determined by an intricate fusion of multimodal information. Therefore, we develop a Self-supervised Multimodal Graph Convolutional Network (SMGCN), which aims to learn the cross-modal user preferences over multiple modalities with an expressive multimodal fusion on a single graph. More importantly, to facilitate and enhance multimodal fusion in …

We propose a method for adding sound-guided visual effects to specific regions of videos with a zero-shot setting. Animating the appearance of the visual effect is challenging because each frame of the edited video should have visual changes while maintaining temporal consistency. Moreover, existing video editing solutions focus on temporal consistency across frames, ignoring the visual style variations over time, e.g., thunderstorm, wave, fire crackling. To overcome this limitation, we utilize temporal sound features for the dynamic style. Specifically, we guide denoising diffusion probabilistic models with an audio latent representation in the audio-visual latent space. To the best of our knowledge, our work is the first to explore sound-guided natural video editing from various sound sources with sound-specialized properties, such as intensity, timbre, and volume. Additionally, we design optical flow-based guidance to generate temporally consistent video frames, capturing the pixel-wise relationship between adjacent frames. Experimental results show that our method outperforms existing video editing techniques, producing more realistic visual effects that reflect the properties of sound. Please visit our page: https://kuai-lab.github.io/soundini-gallery/.

We propose a conditional generative model, named Part Geometry Network (PG-Net), which synthesizes realistic objects and can be used as a robust feature descriptor for object reconstruction and classification. Surface and volumetric representations of objects have complementary properties of three-dimensional objects. Combining these modalities is more informative than using one modality alone. Therefore, PG-Net utilizes complementary properties of surface and volumetric representations by estimating curvature, surface area, and occupancy in voxel grids of objects with a single decoder as a multi-task learning. Objects are combinations of multiple parts, and therefore part geometry (PG) is essential to synthesize each part of the objects. PG-Net employs a part identifier to learn the part geometry. Additionally, we augmented a dataset by interpolating individual functional parts such as wings of an airplane …

Current multi-view 3D object detection methods often fail to detect objects in the overlap region properly, and the networks' understanding of the scene is often limited to that of a monocular detection network. Moreover, objects in the overlap region are often largely occluded or suffer from deformation due to camera distortion, causing a domain shift. To mitigate this issue, we propose using the following two main modules: (1) Stereo Disparity Estimation for Weak Depth Supervision and (2) Adversarial Overlap Region Discriminator. The former utilizes the traditional stereo disparity estimation method to obtain reliable disparity information from the overlap region. Given the disparity estimates as supervision, we propose regularizing the network to fully utilize the geometric potential of binocular images and improve the overall detection accuracy accordingly. Further, the latter module minimizes the representational gap between non-overlap and overlapping regions. We demonstrate the effectiveness of the proposed method with the nuScenes large-scale multi-view 3D object detection data. Our experiments show that our proposed method outperforms current state-of-the-art models, i.e., DETR3D and BEVDet.

While recent 3D instance segmentation approaches show promising results based on transformer architectures they often fail to correctly identify instances with similar appearances. They also ambiguously determine edges leading to multiple misclassifications of adjacent edge points. In this work we introduce a novel framework called EASE to overcome these challenges and improve the perception of complex 3D instances. We first propose a semantic guidance network to leverage rich semantic knowledge from a language model as intelligent priors enhancing the functional understanding of real-world instances beyond relying solely on geometrical information. We explicitly instruct the basic instance queries using text embeddings of each instance to learn deep semantic details. Further we utilize the edge prediction module encouraging the segmentation network to be edge-aware. We extract voxel-wise edge maps from point features and use them as auxiliary information for learning edge cues. In our extensive experiments on large-scale benchmarks ScanNetV2 ScanNet200 S3DIS and STPLS3D our EASE outperforms existing state-of-the-art models demonstrating its superior performance.

Recent LiDAR-based 3D Object Detection (3DOD) methods show promising results, but they often do not generalize well to target domains outside the source (or training) data distribution. To reduce such domain gaps and thus to make 3DOD models more generalizable, we introduce a novel unsupervised domain adaptation (UDA) method, called CMDA, which (i) leverages visual semantic cues from an image modality (i.e., camera images) as an effective semantic bridge to close the domain gap in the cross-modal Bird's Eye View (BEV) representations. Further, (ii) we also introduce a self-training-based learning strategy, wherein a model is adversarially trained to generate domain-invariant features, which disrupt the discrimination of whether a feature instance comes from a source or an unseen target domain. Overall, our CMDA framework guides the 3DOD model to generate highly informative and domain-adaptive features for novel data distributions. In our extensive experiments with large-scale benchmarks, such as nuScenes, Waymo, and KITTI, those mentioned above provide significant performance gains for UDA tasks, achieving state-of-the-art performance.

Computer Vision Lab

교수님 소개

김상필

연구실 소개

연구실 키워드

알럼나이가 간 회사

연구실 지원 방법

대우 조건

연도별 논문 인용수

저널 & 학회 논문 출판 수

연구실 논문 목록

Egocentric Hand Activity Video Dataset and Bidirectional Motion-Priors for Hand Action Recognition

Vision-Guided Targeted Grasping and Vibration for Robotic Pollination in Controlled Environments

3d occupancy prediction with low-resolution queries via prototype-aware view transformation

연구실 보유 특허

대학별 입시끈

우리 연구실 모집

교수님 소개

김상필

연구실 키워드

알럼나이가 간 회사

연구실 지원 방법

대우 조건

연도별 논문 인용수

저널 & 학회 논문 출판 수

연구실 논문 목록

Egocentric Hand Activity Video Dataset and Bidirectional Motion-Priors for Hand Action Recognition

Vision-Guided Targeted Grasping and Vibration for Robotic Pollination in Controlled Environments

3d occupancy prediction with low-resolution queries via prototype-aware view transformation

연구실 보유 특허

최근 3년간 연구실적 높은 연구실

대학별 입시끈

우리 연구실 모집