This work has been submitted to the IEEE for possible publication

Oct 13, 2025 by vinooalluri.com 86 views

Latest 15 Papers - March 10, 2025

6d Object Pose Estimation

6D object pose estimation is a crucial task in computer vision, which involves estimating the 3D pose of an object from a 2D image. This task has numerous applications in robotics, computer-aided design, and augmented reality. In recent years, significant progress has been made in this field, with the development of various deep learning-based methods.

Title	Date	Comment
Active 6D Pose Estimation for Textureless Objects using Multi-View RGB Frames	2025-03-05
Improving 6D Object Pose Estimation of metallic Household and Industry Objects	2025-03-05
Increasing the Task Flexibility of Heavy-Duty Manipulators Using Visual 6D Pose Estimation of Objects	2025-02-26
EfficientPose 6D: Scalable and Efficient 6D Object Pose Estimation	2025-02-19
Enhancing Transparent Object Pose Estimation: A Fusion of GDR-Net and Edge Detection	2025-02-17	accep... accepted at First Austrian Symposium on AI, Robotics, and Vision (AIROV 2024)
HIPPo: Harnessing Image-to-3D Priors for Model-free Zero-shot 6D Pose Estimation	2025-02-14
CordViP: Correspondence-based Visuomotor Policy for Dexterous Manipulation in Real-World	2025-02-12
Advanced Object Detection and Pose Estimation with Hybrid Task Cascade and High-Resolution Networks	2025-02-06
FreeZe: Training-free zero-shot 6D pose estimation with geometric and vision foundation models	2025-01-08	Accep... Accepted to ECCV 2024. Project page: https://andreacaraffa.github.io/freeze
A Novel Convolution and Attention Mechanism-based Model for 6D Object Pose Estimation	2024-12-31	8 pag... 8 pages, 4 figures, 6 tables
ReFlow6D: Refraction-Guided Transparent Object 6D Pose Estimation via Intermediate Representation Learning	2024-12-30
Level-Set Parameters: Novel Representation for 3D Shape Analysis	2024-12-18
KITchen: A Real-World Benchmark and Dataset for 6D Object Pose Estimation in Kitchen Environments	2024-12-17	This ... This work has been accepted for publishing at The 2024 IEEE-RAS International Conference on Humanoid Robots
COBRA -- COnfidence score Based on shape Regression Analysis for method-independent quality assessment of object pose estimation from single images	2024-12-16
Category Level 6D Object Pose Estimation from a Single RGB Image using Diffusion	2024-12-16

Human Pose Estimation

Human pose estimation is a fundamental task in computer vision, which involves estimating the 3D pose of a human from a 2D image. This task has numerous applications in robotics, computer-aided design, and augmented reality. In recent years, significant progress has been made in this field, with the development of various deep learning-based methods.

Title	Date	Comment
Refinement Module based on Parse Graph of Feature Map for Human Pose Estimation	2025-03-02
BGM2Pose: Active 3D Human Pose Estimation with Non-Stationary Sounds	2025-03-01
Transformers with Joint Tokens and Local-Global Attention for Efficient Human Pose Estimation	2025-02-28	This ... This work has been submitted to the IEEE for possible publication
BST: Badminton Stroke-type Transformer for Skeleton-based Action Recognition in Racket Sports	2025-02-28	8 pag... 8 pages (excluding references). The code will be released in a few months
Sixth-Sense: Self-Supervised Learning of Spatial Awareness of Humans from a Planar Lidar	2025-02-28
STGFormer: Spatio-Temporal GraphFormer for 3D Human Pose Estimation in Video	2025-02-26
Pose Magic: Efficient and Temporally Consistent Human Pose Estimation with a Hybrid Mamba-GCN Network	2025-02-26	This ... This work has been accepted by AAAI 2025
EgoSim: An Egocentric Multi-view Simulator and Real Dataset for Body-worn Cameras during Motion and Activity	2025-02-25
Leveraging 2D Masked Reconstruction for Domain Adaptation of 3D Pose Estimation	2025-02-25	16 pages, 7 figures
CHAMP: Conformalized 3D Human Multi-Hypothesis Pose Estimators	2025-02-23
DeProPose: Deficiency-Proof 3D Human Pose Estimation via Adaptive Multi-View Fusion	2025-02-23	The s... The source code will be available at https://github.com/WUJINHUAN/DeProPose
Pose Prior Learner: Unsupervised Categorical Prior Learning for Pose Estimation	2025-02-20
VarGes: Improving Variation in Co-Speech 3D Gesture Generation via StyleCLIPS	2025-02-18
Spatiotemporal Multi-Camera Calibration using Freely Moving People	2025-02-18	8 pages, 4 figures
X-Fi: A Modality-Invariant Foundation Model for Multimodal Human Sensing	2025-02-17

Gaussian Splatting

Gaussian splatting is a technique used in computer vision to represent 3D scenes as a collection of Gaussian distributions. This technique has numerous applications in robotics, computer-aided design, and augmented reality. In recent years, significant progress has been made in this field, with the development of various deep learning-based methods.

| Title | **Date
Q&A: Latest 15 Papers - March 10, 2025

Q: What is 6D object pose estimation?

A: 6D object pose estimation is a task in computer vision that involves estimating the 3D pose of an object from a 2D image. This task has numerous applications in robotics, computer-aided design, and augmented reality.

Q: What are some recent advancements in 6D object pose estimation?

A: Recent advancements in 6D object pose estimation include the development of various deep learning-based methods, such as active 6D pose estimation, multi-view RGB frames, and hybrid task cascade and high-resolution networks.

Q: What is human pose estimation?

A: Human pose estimation is a fundamental task in computer vision that involves estimating the 3D pose of a human from a 2D image. This task has numerous applications in robotics, computer-aided design, and augmented reality.

Q: What are some recent advancements in human pose estimation?

A: Recent advancements in human pose estimation include the development of various deep learning-based methods, such as refinement modules, transformers with joint tokens and local-global attention, and spatio-temporal graph formers.

Q: What is Gaussian splatting?

A: Gaussian splatting is a technique used in computer vision to represent 3D scenes as a collection of Gaussian distributions. This technique has numerous applications in robotics, computer-aided design, and augmented reality.

Q: What are some recent advancements in Gaussian splatting?

A: Recent advancements in Gaussian splatting include the development of various deep learning-based methods, such as multimodal place recognition, efficient video representation and compression, and sparse-view super-resolution 3D Gaussian splatting.

Q: What is diffusion?

A: Diffusion is a technique used in computer vision to represent 3D scenes as a collection of Gaussian distributions. This technique has numerous applications in robotics, computer-aided design, and augmented reality.

Q: What are some recent advancements in diffusion?

A: Recent advancements in diffusion include the development of various deep learning-based methods, such as fluid reconstruction and prediction, coarse graining and reduced order models, and compositional world knowledge.

Q: How can I stay up-to-date with the latest advancements in these fields?

A: You can stay up-to-date with the latest advancements in these fields by following reputable sources, such as arXiv, ResearchGate, and Academia.edu. You can also attend conferences and workshops, and participate in online forums and discussions.

Q: What are some potential applications of these techniques?

A: These techniques have numerous potential applications in robotics, computer-aided design, and augmented reality. They can be used to improve the accuracy and efficiency of object recognition, pose estimation, and scene understanding.

Q: What are some potential challenges and limitations of these techniques?

A: These techniques have several potential challenges and limitations, including the need for large amounts of training data, the risk of overfitting, and the difficulty of generalizing to new scenarios.

Q: How can I get started with implementing these techniques?

A: You can get started with implementing these techniques by following online tutorials and documentation, and by participating in online forums and discussions. You can also join research groups and collaborate with other researchers to gain experience and expertise.

Q: What are some potential future directions for these techniques?

A: Some potential future directions for these techniques include the development of more efficient and accurate methods, the integration of multiple modalities, and the application of these techniques to new domains and scenarios.

Q: How can I contribute to the development of these techniques?

A: You can contribute to the development of these techniques by participating in online forums and discussions, by sharing your own research and ideas, and by collaborating with other researchers to advance the state-of-the-art.