Search Results for author:
Found 21 papers, 9 papers with code
3DGAA: Realistic and Robust 3D Gaussian-based Adversarial Attack for Autonomous Driving
In this work, we propose 3D Gaussian-based Adversarial Attack (3DGAA), a novel adversarial object generation framework that leverages the full 14-dimensional parameterization of 3D Gaussian Splatting (3DGS) to jointly optimize geometry and appearance in physically realizable ways.
L2HCount:Generalizing Crowd Counting from Low to High Crowd Density via Density Simulation
Third, we propose a Dual-Density Memory Encoding Module that uses two crowd memories to learn scene-specific patterns from low- and simulated high-density scenes, respectively.
GaussianCAD: Robust Self-Supervised CAD Reconstruction from Three Orthographic Views Using 3D Gaussian Splatting
While this reformulation offers a promising perspective, existing 3D reconstruction methods typically require natural images and corresponding camera poses as inputs, which introduces two major significant challenges: (1) modality discrepancy between CAD sketches and natural images, and (2) difficulty of accurate camera pose estimation for CAD sketches.
QUART-Online: Latency-Free Large Multimodal Language Model for Quadruped Robot Learning
This paper addresses the inherent inference latency challenges associated with deploying multimodal large language models (MLLM) in quadruped vision-language-action (QUAR-VLA) tasks.
MamKPD: A Simple Mamba Baseline for Real-Time 2D Keypoint Detection
Moreover, MamKPD achieves state-of-the-art results on the MPII dataset and competitive results on the AP-10K dataset while saving 85% of the parameters compared to ViTPose.
Towards Physically Realizable Adversarial Attacks in Embodied Vision Navigation
To make the patch inconspicuous to human observers, we introduce a two-stage opacity optimization mechanism, in which opacity is fine-tuned after texture optimization.
ActivityCLIP: Enhancing Group Activity Recognition by Mining Complementary Information from Text to Supplement Image Modality
Thus, we propose ActivityCLIP, a plug-and-play method for mining the text information contained in the action labels to supplement the image information for enhancing group activity recognition.
Micro-expression recognition based on depth map to point cloud
Conversely, depth information can effectively represent motion information related to facial structure changes and is not affected by lighting.
DHRNet: A Dual-Path Hierarchical Relation Network for Multi-Person Pose Estimation
Multi-person pose estimation (MPPE) presents a formidable yet crucial challenge in computer vision.
Full-frequency dynamic convolution: a physical frequency-dependent convolution for sound event detection
Recently, 2D convolution has been found unqualified in sound event detection (SED).
Spatial-Temporal Decoupling Contrastive Learning for Skeleton-based Human Action Recognition
Furthermore, to explicitly exploit the latent data distributions, we employ the attentive features to contrastive learning, which models the cross-sequence semantic relations by pulling together the features from the positive pairs and pushing away the negative pairs.
BiHRNet: A Binary high-resolution network for Human Pose Estimation
On the challenging of COCO dataset, the proposed method enables the binary neural network to achieve 70. 8 mAP, which is better than most tested lightweight full-precision networks.
SiT-MLP: A Simple MLP with Point-wise Topology Feature Learning for Skeleton-based Action Recognition
Graph convolution networks (GCNs) have achieved remarkable performance in skeleton-based action recognition.
Physics-constrained Attack against Convolution-based Human Motion Prediction
Specifically, we introduce a novel adaptable scheme that facilitates the attack to suit the scale of the target pose and two physical constraints to enhance the naturalness of the adversarial example.
An Improved Baseline Framework for Pose Estimation Challenge at ECCV 2022 Visual Perception for Navigation in Human Environments Workshop
This technical report describes our first-place solution to the pose estimation challenge at ECCV 2022 Visual Perception for Navigation in Human Environments Workshop.
Leveraging the Video-level Semantic Consistency of Event for Audio-visual Event Localization
In contrast to existing methods, we propose a novel video-level semantic consistency guidance network for the AVE localization task.
Kinematics Modeling Network for Video-based Human Pose Estimation
In this work, we propose a plug-and-play kinematics modeling module (KMM) to explicitly model temporal correlations between joints across different frames by calculating their temporal similarity.
Learning Constrained Dynamic Correlations in Spatiotemporal Graphs for Motion Prediction
The key of DSTD-GC is constrained dynamic correlation modeling, which explicitly parameterizes the common static constraints as a spatial/temporal vanilla adjacency matrix shared by all frames/joints and dynamically extracts correspondence variances for each frame/joint with an adjustment modeling function.
Relation-Based Associative Joint Location for Human Pose Estimation in Videos
Moreover, the JRE can infer invisible joints according to the relationship between joints, which is beneficial for the model to locate occluded joints.
Energy-based Periodicity Mining with Deep Features for Action Repetition Counting in Unconstrained Videos
Our work features two-fold: 1) An important insight that deep features extracted for action recognition can well model the self-similarity periodicity of the repetitive action is presented.
DWnet: Deep-Wide Network for 3D Action Recognition
We propose in this paper a deep-wide network (DWnet) which combines the deep structure with the broad learning system (BLS) to recognize actions.
