New architecture for RGB-D object recognition outperforms existing methods.
problem Improving object recognition in RGB-D cameras.
method RCFusion architecture combining RGB and depth information.
result RCFusion outperforms existing methods on standard datasets.
Scene recognition with RGB images has been extensively studied and has reached very remarkable recognition levels, thanks to convolutional neural networks (CNN) and large scene datasets. In contrast, current RGB-D scene data is much more limited, so often leverages RGB large datasets, by transferring pretrained RGB CNN…
A new data-level recombination strategy improves RGB-D salient object detection.
problem RGB-D salient object detection struggles with depth information.
method Proposes a novel data-level recombination strategy to fuse RGB and depth data before feature extraction.
result Achieves a new state-of-the-art performance in RGB-D salient object detection.
Study presents a low-cost local motion planner for vineyard navigation.
problem Autonomous navigation in vineyards with limited resources.
method RGB-D camera, dual layer control algorithm, deep learning synergy.
result Robust motion planning for vineyard navigation achieved.
ObSuRF converts a single image into a 3D model with NeRFs.
problem Creating a 3D model from a single image with object segmentation.
method Unsupervised volume segmentation using Neural Radiance Fields (NeRFs).
result ObSuRF can segment a 3D scene into objects from a single image.
New dataset and benchmarks for lifelong robotic vision tasks.
problem Challenges in applying computer vision to robots, especially lifelong learning.
method Provided a new lifelong robotic vision dataset and benchmarks.
result Demonstrated the complexity and bottlenecks in lifelong object recognition.
Project aims to improve robot vision by adapting to new environments.
problem Robots struggle to recognize objects in new settings.
method Selected and evaluated state-of-the-art DA methods on various datasets.
result Training a good object classifier remains challenging.
The introduction of cheap RGB-D cameras, stereo cameras, and LIDAR devices has given the computer vision community 3D information that conventional RGB cameras cannot provide. This data is often stored as a point cloud. In this paper, we present a novel method to apply the concept of convolutional neural networks to th…
STRING improves 2D and 3D position encodings for better performance.
problem Efficient and accurate position encoding for 2D and 3D applications.
method STRING extends Rotary Position Encodings with a unifying theoretical framework, maintaining translation invariance and low computational cost.
result STRING shows substantial gains in open-vocabulary object detection and robotics.
A new method classifies signals on irregular domains using convolutional cluster pooling.
problem Classifying signals on irregular domains with shared properties.
method Convolutional Cluster Pooling layer exploiting multi-scale clustering.
result Generalizes CNNs for graph data, capturing both local and global patterns.
SLAM-net learns to navigate visually in challenging indoor environments.
problem Challenges in SLAM for visual robot navigation, especially in noisy conditions.
method Differentiable SLAM Network (SLAM-net) that encodes a particle filter SLAM algorithm in a differentiable graph and learns components through backpropagation.
result Significantly outperforms ORB-SLAM in noisy conditions and improves the Habitat Challenge 2020 PointNav task.
Regularizes deep multi-task networks to prevent task interference.
problem Interfering tasks in deep neural networks reduce overall performance.
method Proposes a gradient regularization term to minimize task interference.
result Models with orthogonal gradients perform better on various datasets.
3D object recognition accuracy can be improved by learning the multi-scale spatial features from 3D spatial geometric representations of objects such as point clouds, 3D models, surfaces, and RGB-D data. Current deep learning approaches learn such features either using structured data representations (voxel grids and o…
Paper combines RNN and signatures for learning functions on streamed multimodal data.
problem Learning functions on streamed multimodal data.
method Hybrid Logsig-RNN algorithm combining signatures and RNN.
result Hybrid algorithm achieves outstanding accuracy with superior efficiency and robustness.
ES-VAE models skeletal pose trajectories by removing nuisance factors.
problem Handling camera orientation, subject scale, viewpoint, and execution speed in skeletal data.
method ES-VAE uses TSRVF representation on Kendall's shape manifold to isolate shape dynamics.
result ES-VAE outperforms standard VAEs and sequence modeling baselines in gait cycle prediction and action recognition.