55 accepted papers
Ground-air-based visual localization for UAVs, which typically relies on a single camera to track artificial markers, is often hampered by a limited field of view and sensitivity to illumination. Although active camera scheduling can mitigate these issues to some extent, it often incurs high motion
In recent years, parametric representations of point clouds have been widely applied in tasks such as memory-efficient mapping and multi-robot collaboration. Highly adaptive models, like spline surfaces or quadrics, are computationally expensive in detection or fitting. In contrast, real-time method…
The rapid development and widespread adoption of Audio Large Language Models (ALLMs) require a rigorous assessment of their trustworthiness. However, existing evaluation frameworks, primarily designed for text, are not equipped to handle the unique vulnerabilities introduced by audio’s acoustic prop…
Differentially private (DP) image synthesis enables the generation of realistic images while bounding privacy leakage, facilitating secure data sharing across organizations. However, the Gaussian noise injected during DP training, such as via DP-SGD, often severely degrades synthesis quality by disr
While edge GPUs are increasingly used for latency-critical DNN tasks, limited resources often fail to meet strict real-time (RT) requirements under concurrent workloads. Existing preemption and early-exit mechanisms often underutilize GPU resources through single-task queuing and sacrifice excessive…
Reflective imaging enables the mirror imagings and physical entities to possess identical attributes, e.g., color and shape. Current mirror detection (MD) methods primarily rely on designing functional components to establish the correlation and disparities between the imagings and entities, thereby
With the rapid growth of the Internet of Things (IoT), integrating artificial intelligence (AI) on extremely weak embedded devices has garnered significant attention, enabling improved real-time performance and enhanced data privacy. However, the resource limitations of such devices and unreliable n
We present SHARP, an approach to photorealistic view synthesis from a single image. Given a single photograph, SHARP regresses the parameters of a 3D Gaussian representation of the depicted scene. This is done in less than a second on a standard GPU via a single feedforward pass through a neural net…
Real-world dark images commonly exhibit not only low visibility and contrast but also complex noise and blur, posing significant restoration challenges. Existing methods often rely on paired data or fail to model dynamic illumination and blur characteristics, leading to poor generalization. To tackl
Recurrent Neural Networks (RNNs) are widely used in control system due to their dynamic capabilities. However, the control accuracy of RNN-based systems can be compromised by noise interference, and there has been little research on RNN-based control in disturbed multi-agent systems. To address this…
This paper presents the design and development of a jet vectoring backpack (JVP) - a single-person vertical take-off and landing (VTOL) aircraft specifically designed for emergency and disaster relief. Unlike existing electrically powered personal flying devices, the JVP employs turbojet engines as
Large language models are reshaping internet services. Serving these models is often costly, as it requires multiple high-end GPUs. Consumer-grade GPUs offer cheaper computational power, providing an opportunity for more cost-efficient LLM serving.Prior efforts have explored distributed serving at s
A prevalent approach in Parameter-Efficient Fine-Tuning (PEFT) of pre-trained Vision Transformers (ViT) involves freezing the majority of the backbone parameters and solely learning low-rank adaptation weight matrices to accommodate downstream tasks. These low-rank matrices are commonly derived thro…
Federated Learning (FL) studies often assume a static data distribution, whereas real-world scenarios involve dynamic changes. To address this gap, we study Federated Continuous Category Discovery and Learning (FC^2DL), an essential yet underexplored problem that enables FL models to evolve continuo…
Low-Light Image Enhancement (LLIE) is a crucial computer vision task that aims to restore detailed visual information from corrupted low-light images. Many existing LLIE methods are based on standard RGB (sRGB) space, which often produce color bias and brightness artifacts due to inherent high color…
The increasing prevalence of embedded systems has necessitated manufacturers to migrate product code, transferring existing products to new embedded operating systems (OSes) for getting better compatibility and performance. Since manufacturers’ product code predominantly employs the Thing Specificat
Analyzing and forecasting trajectories of agents like pedestrians plays a pivotal role for embodied intelligent applications. The inherent indeterminacy of human behavior and complex social interaction among a rich variety of agents makes this task more challenging than common time-series forecastin
Turbojet-powered VTOL autonomous aerial vehicle (AAVs) have garnered increased attention in heavy-load transport and emergency services, due to their superior power density and thrust-to-weight ratio compared to existing electronic propulsion systems. The main challenge with jet-powered AAVs lies in
The rapid advancement of large language models (LLMs) in recent years has made it feasible to establish domain-specific LLMs for specialized fields. However, in practical development, acquiring domain-specific knowledge often requires a significant amount of professional expert manpower. Moreover, e…
Recent interactive matting methods have demonstrated satisfactory performance in capturing the primary regions of objects, but they fall short in extracting fine-grained details in edge regions. Diffusion models trained on billions of image-text pairs, demonstrate exceptional capability in modeling…
Vision Transformers (ViTs), extensively pre-trained on large-scale datasets, have become fundamental to foundation models, enabling adaptation to diverse downstream tasks. Existing adaptation methods typically require direct data access, rendering them infeasible in privacy-sensitive domains where c…
People have diverse privacy requirements. This is best modeled using a personalized local differential privacy model where each user privatizes their data using a possibly different privacy parameter. While the model of personalized local differential privacy is a natural and important one, prior w…
Currently, inspired by the success of vision-language models (VLMs), an increasing number of researchers are focusing on improving VLMs and have achieved promising results. However, most existing methods concentrate on optimizing the connector and enhancing the language model component, while neglec…
For a swarm of Unmanned Aerial Vehicle (UAV), long-distance visual mapping is advantageous for pre-planning navigation paths in unknown urban building environments. This work leverages two cameras on two UAVs to build a wide-baseline collaborative stereo camera, which can construct a navigable mesh…
We introduce BEVRender, a novel learning-based approach for the localization of ground vehicles in Global Navigation Satellite System (GNSS)-denied off-road scenarios. These environments are typically challenging for conventional vision-based state estimation due to the lack of distinct visual landm…
Exposure Correction (EC) aims to recover proper exposure conditions for images captured under over-exposure or under-exposure scenarios. While existing deep learning models have shown promising results, few have fully embedded Retinex theory into their architecture, highlighting a gap in current met…
A common strategy for Parameter-Efficient Fine-Tuning (PEFT) of pre-trained Vision Transformers (ViTs) involves adapting the model to downstream tasks by learning a low-rank adaptation matrix. This matrix is decomposed into a product of down-projection and up-projection matrices, with the bottleneck…
Current formations commonly rely on invariant hierarchical structures, such as predetermined leaders or enumerated formation shapes. These structures could be unidirectional and sluggish, constraining their flexibility and agility when encountering cluttered environments. To surmount these constrain
"Most existing Low-light Image Enhancement (LLIE) methods either directly map Low-Light (LL) to Normal-Light (NL) images or use semantic or illumination maps as guides. However, the ill-posed nature of LLIE and the difficulty of semantic retrieval from impaired inputs limit these methods, especially…
Cable-driven exosuits have the advantages of being lightweight, compact, and well-adapted due to avoiding extensive usage of rigid structures to transmit power. However, applying too many flexible materials into the system will cause some problems such as insufficient stiffness, easy deformation, an
Parameter-efficient fine-tuning for pre-trained Vision Transformers aims to adeptly tailor a model to downstream tasks by learning a minimal set of new adaptation parameters while preserving the frozen majority of pre-trained parameters. Striking a balance between retaining the generalizable represe…
Neural architecture search (NAS) has shown great promise in automatically designing lightweight models. However, conventional approaches are insufficient in training the supernet and pay little attention to actual robot hardware resources. To meet such challenges, we propose RAM-NAS, a resource-awar…
This letter presents a sample-efficient deep reinforcement learning (DRL) based method to address the intricate task of autonomous navigation in dynamic, constrained environments. The proposed method leverages a graph neural network with feature-wise linear modulation modules to effectively extract
To achieve effective collaboration of multiple robots, it requires efficient exchanges of map information. As directly exchanging generally used depth map requires high communication bandwidth, it is practical to enhance the efficiency using map compression techniques based on Gaussian mixture model
The advent of high-capacity pre-trained models has revolutionized problem-solving in computer vision, shifting the focus from training task-specific models to adapting pre-trained models. Consequently, effectively adapting large pre-trained models to downstream tasks in an efficient manner has becom…
Indoor scene reconstruction from monocular images has long been sought after by augmented reality and robotics developers. Recent advances in neural field representations and monocular priors have led to remarkable results in scene-level surface reconstructions. The reliance on Multilayer Perceptron…
Autonomous navigation of Micro Aerial Vehicles (MAVs) in dynamic and unknown environments is a complex and challenging task. Current works rely on assumptions to solve the problem. The MAV's pose is precisely known, the dynamic obstacles can be explicitly segmented from static ones, their number is
The vision-based relative localization can provide effective feedback for the cooperation of aerial swarm and has been widely investigated in previous works. However, the limited field of view (FOV) inherently restricts its performance. To cope with this issue, we propose a novel distributed active
In this paper, we present two new algorithms for covariance estimation under concentrated differential privacy (zCDP). The first algorithm achieves a Frobenius error of $\tilde{O}(d^{1/4}\sqrt{\mathrm{tr}}/\sqrt{n} + \sqrt{d}/n)$, where $\mathrm{tr}$ is the trace of the covariance matrix. By takin…
3D imaging radar offers robust perception capability through visually demanding environments due to the unique penetrative and reflective properties of millimeter waves (mmWave). Current approaches for 3D perception with imaging radar require knowledge of environment geometry, accumulation of data f…
The key towards learning informative node representations in graphs lies in how to gain contextual information from the neighbourhood. In this work, we present a simple-yet-effective self-supervised node representation learning strategy via directly maximizing the mutual information between the hidd…
This paper proposes a perception-shared and swarm trajectory global optimal (STGO) algorithm fused UAVs formation motion planning framework aided by an active sensing system. First, the point cloud received by each UAV is fit by the gaussian mixture model (GMM) and transmitted in the swarm. Resampli…
We present a fast, scalable, and accurate Simultaneous Localization and Mapping (SLAM) system that represents indoor scenes as a graph of objects. Leveraging the observation that artificial environments are structured and occupied by recognizable objects, we show that a compositional and scalable ob…
Quadrotor with active sensing was proposed recently to overcome the view field limitation and achieved an excellent perception ability in obstacle avoidance tasks. To realize high-speed flights of this quadrotor in unknown and cluttered environments, a computationally efficient trajectory planner is
We aim to assess the performance of LiDAR-to-map registration on compressive maps. Modern autonomous vehicles utilize pre-built HD (High-Definition) maps to perform sensor-to-map registration, which recovers pose estimation failures and reduces drift in a large-scale environment. However, sensor-to-…
We present self-supervised geometric perception (SGP), the first general framework to learn a feature descriptor for correspondence matching without any ground-truth geometric model labels (e.g., camera poses, rigid transformations). Our first contribution is to formulate geometric perception as an…
To perch on or grasp the objective surface using the suction cup-based manipulator, the precise contact control is commonly required. Improper contact angle or insufficient contact force may cause failure. To enhance the tolerance to flight control insufficiency, a suction cup that comprises an inne
We present Deep Global Registration, a differentiable framework for pairwise registration of real-world 3D scans. Deep global registration is based on three modules: a 6-dimensional convolutional network for correspondence confidence prediction, a differentiable Weighted Procrustes algorithm for clo…
We propose a fast and accurate 3D reconstruction system that takes a sequence of RGB-D frames and produces a globally consistent camera trajectory and a dense 3D geometry. We redesign core modules of a state-of-the-art offline reconstruction pipeline to maximally exploit the power of GPU. We introdu…
Achieving high surface reconstruction accuracy in dense mapping has been a desirable target for both robotics and vision communities. In the robotics literature, simultaneous localization and mapping (SLAM) systems use RGB-D cameras to reconstruct a dense map of the environment. They leverage the de…
Mesh plays an indispensable role in dense realtime reconstruction essential in robotics. Efforts have been made to maintain flexible data structures for 3D data fusion, yet an efficient incremental framework specifically designed for online mesh storage and manipulation is missing. We propose a nove…
We propose a novel 3D spatial representation for data fusion and scene reconstruction. Probabilistic Signed Distance Function (Probabilistic SDF, PSDF) is proposed to depict uncertainties in the 3D space. It is modeled by a joint distribution describing SDF value and its inlier probability, reflecti…
The kinematic modelling has been applied to many controllers of under-actuated manipulators. Most of these studies assume that the control process is conducted within the workspace. However, as such a kinematic model cannot describe the situations when the stable grasping is violated in the real env…
This paper presents a trajectory tracking control strategy based on the subspace stabilization approach to accurately manipulate an under-actuated flying robot from a known initial state to the desired terminal state. To facilitate the development of this tracking strategy, the dynamical model of th…
In robot-assisted upper limb rehabilitation, detecting the intentions of hemiplegic patients is essential towards assisting the patients to actively exercise instead of driving passive motions. Many interactive channels, such as voice, EMG and EEG, have been studied to estimate the motion intentions…