Depth-guided Free-space Segmentation for a Mobile Robot
Not on the map. Why?
Abstract (arXiv)
Accurate indoor free-space segmentation is a challenging task due to the complexity and the dynamic nature that indoor environments exhibit. We propose an indoors free-space segmentation method that associates large depth values with navigable regions. Our method leverages an unsupervised masking technique that, using positive instances, generates segmentation labels based on textural homogeneity and depth uniformity. Moreover, we generate superpixels corresponding to areas of higher depth and align them with features extracted from a Dense Prediction Transformer (DPT). Using the estimated free-space masks and the DPT feature representation, a SegFormer model is fine-tuned on our custom-collected indoor dataset. Our experiments demonstrate sufficient performance in intricate scenarios characterized by cluttered obstacles and challenging identification of free space.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 26 of its 31 references that are in the map. First posted on arXiv 3 Nov 2023.
Builds on
26 of 31 references.
References · 26
- 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
- 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
- 2017Mask R-CNNcited by 7,259
- 2021SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformerscited by 2,923
- 2021Vision Transformers for Dense Predictioncited by 1,789
- 2014Depth Map Prediction from a Single Image using a Multi-Scale Deep Networkcited by 1,264
- 2020Object-Centric Learning with Slot Attentioncited by 865
- 2021Uformer: A General U-Shaped Transformer for Image Restorationcited by 864
- 2021Segmenter: Transformer for Semantic Segmentationcited by 855
- 2017Unsupervised Learning of Depth and Ego-Motion from Videocited by 778
- 2021TransFuse: Fusing Transformers and CNNs for Medical Image Segmentationcited by 258
- 2021Conditional Object-Centric Learning from Videocited by 247
14 more
- 2018Learning from positive and unlabeled data: a surveycited by 218
- 2018DF-Net: Unsupervised Joint Learning of Depth and Flow using Cross-Task Consistencycited by 148
- 2018Depth-aware CNN for RGB-D Segmentationcited by 73
- 2019Deep Visual MPC-Policy Learning for Navigationcited by 54
- 2018Real-Time Joint Semantic Segmentation and Depth Estimation Using Asymmetric Annotationscited by 51
- 2018Deep Object-Centric Policies for Autonomous Drivingcited by 41
- 2016Joint Semantic Segmentation and Depth Estimation with Deep Convolutional Networkscited by 36
- 2018GONet: A Semi-Supervised Deep Learning Approach For Traversability Estimationcited by 28
- 2020Perceive, Attend, and Drive: Learning Spatial Attention for Safe Self-Drivingcited by 24
- 2016Exploiting Depth from Single Monocular Images for Object Detection and Semantic Segmentationcited by 10
- 2018VUNet: Dynamic Scene View Synthesis for Traversability Estimation using an RGB Cameracited by 7
- 2022A Survey of Traversability Estimation for Mobile Robotscited by 7
- 2019Enhanced free space detection in multiple lanes based on single CNN with scene identificationcited by 4
- 2017Minimizing Supervision for Free-space Segmentationcited by 3
Built on by
Citing papers per year: 2025: 1.
1 paper.