Trans4Trans: Efficient Transformer for Transparent Object Segmentation to Help Visually Impaired People Navigate in the Real World
arXiv:2107.03172
Abstract
Common fully glazed facades and transparent objects present architectural barriers and impede the mobility of people with low vision or blindness, for instance, a path detected behind a glass door is inaccessible unless it is correctly perceived and reacted. However, segmenting these safety-critical objects is rarely covered by conventional assistive technologies. To tackle this issue, we construct a wearable system with a novel dual-head Transformer for Transparency (Trans4Trans) model, which is capable of segmenting general and transparent objects and performing real-time wayfinding to assist people walking alone more safely. Especially, both decoders created by our proposed Transformer Parsing Module (TPM) enable effective joint learning from different datasets. Besides, the efficient Trans4Trans model composed of symmetric transformer-based encoder and decoder, requires little computational expenses and is readily deployed on portable GPUs. Our Trans4Trans model outperforms state-of-the-art methods on the test sets of Stanford2D3D and Trans10K-v2 datasets and obtains mIoU of 45.13% and 75.14%, respectively. Through various pre-tests and a user study conducted in indoor and outdoor scenarios, the usability and reliability of our assistive system have been extensively verified.
arXiv commentAccepted to ICCV 2021 ACVR Workshop. Code will be made publicly available at https://github.com/jamycheung/Trans4Trans
References in corpus (10)
- ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation
- DUNet: A deformable network for retinal vessel segmentation
- Joint 2D-3D-Semantic Data for Indoor Scene Understanding
- Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
- ResT: An Efficient Transformer for Visual Recognition
- Polarization-driven Semantic Segmentation via Efficient Attention-bridged Fusion
- Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions
- Fully Transformer Networks for Semantic Image Segmentation
- FakeMix Augmentation Improves Transparent Object Detection
- Floor extraction and door detection for visually impaired guidance