[HTML][HTML] A survey of state-of-the-art on visual SLAM
This paper is an overview to Visual Simultaneous Localization and Map** (V-SLAM). We
discuss the basic definitions in the SLAM and vision system fields and provide a review of …
discuss the basic definitions in the SLAM and vision system fields and provide a review of …
A comprehensive survey of visual slam algorithms
Simultaneous localization and map** (SLAM) techniques are widely researched, since
they allow the simultaneous creation of a map and the sensors' pose estimation in an …
they allow the simultaneous creation of a map and the sensors' pose estimation in an …
Panoptic neural fields: A semantic object-aware neural scene representation
We present PanopticNeRF, an object-aware neural scene representation that decomposes
a scene into a set of objects (things) and background (stuff). Each object is represented by a …
a scene into a set of objects (things) and background (stuff). Each object is represented by a …
Visual language maps for robot navigation
Grounding language to the visual observations of a navigating agent can be performed
using off-the-shelf visual-language models pretrained on Internet-scale data (eg, image …
using off-the-shelf visual-language models pretrained on Internet-scale data (eg, image …
Bundlesdf: Neural 6-dof tracking and 3d reconstruction of unknown objects
We present a near real-time (10Hz) method for 6-DoF tracking of an unknown object from a
monocular RGBD video sequence, while simultaneously performing neural 3D …
monocular RGBD video sequence, while simultaneously performing neural 3D …
Navigating to objects in the real world
Semantic navigation is necessary to deploy mobile robots in uncontrolled environments
such as homes or hospitals. Many learning-based approaches have been proposed in …
such as homes or hospitals. Many learning-based approaches have been proposed in …
vmap: Vectorised object map** for neural field slam
We present vMAP, an object-level dense SLAM system using neural field representations.
Each object is represented by a small MLP, enabling efficient, watertight object modelling …
Each object is represented by a small MLP, enabling efficient, watertight object modelling …
Cosypose: Consistent multi-view multi-object 6d pose estimation
We introduce an approach for recovering the 6D pose of multiple known objects in a scene
captured by a set of input images with unknown camera viewpoints. First, we present a …
captured by a set of input images with unknown camera viewpoints. First, we present a …