English
Related papers

Related papers: VANP: Learning Where to See for Navigation with Se…

200 papers

Weakly supervised learning of object detection is an important problem in image understanding that still does not have a satisfactory solution. In this paper, we address this problem by exploiting the power of deep convolutional neural…

Computer Vision and Pattern Recognition · Computer Science 2016-12-20 Hakan Bilen , Andrea Vedaldi

Recent embodied navigation approaches leveraging Vision-Language Models (VLMs) demonstrate strong generalization in versatile Vision-Language Navigation (VLN). However, reliable path planning in complex environments remains challenging due…

This paper reports on the development, execution, and open-sourcing of a new robotics course at MIT. The course is a modern take on "Visual Navigation for Autonomous Vehicles" (VNAV) and targets first-year graduate students and senior…

Vision-Language Navigation (VLN) is a challenging task that requires an embodied agent to perform action-level modality alignment, i.e., make instruction-asked actions sequentially in complex visual environments. Most existing VLN agents…

Computer Vision and Pattern Recognition · Computer Science 2022-06-01 Bingqian Lin , Yi Zhu , Zicong Chen , Xiwen Liang , Jianzhuang Liu , Xiaodan Liang

Developments in machine learning interpretability techniques over the past decade have provided new tools to observe the image regions that are most informative for classification and localization in artificial neural networks (ANNs). Are…

Computer Vision and Pattern Recognition · Computer Science 2021-11-02 Thomas A. Langlois , H. Charles Zhao , Erin Grant , Ishita Dasgupta , Thomas L. Griffiths , Nori Jacoby

How much does having visual priors about the world (e.g. the fact that the world is 3D) assist in learning to perform downstream motor tasks (e.g. navigating a complex environment)? What are the consequences of not utilizing such visual…

Computer Vision and Pattern Recognition · Computer Science 2019-12-25 Alexander Sax , Jeffrey O. Zhang , Bradley Emi , Amir Zamir , Silvio Savarese , Leonidas Guibas , Jitendra Malik

We propose to take a novel approach to robot system design where each building block of a larger system is represented as a differentiable program, i.e. a deep neural network. This representation allows for integrating algorithmic planning…

Robotics · Computer Science 2018-07-19 Peter Karkus , David Hsu , Wee Sun Lee

Recently, data-driven deep saliency models have achieved high performance and have outperformed classical saliency models, as demonstrated by results on datasets such as the MIT300 and SALICON. Yet, there remains a large gap between the…

Computer Vision and Pattern Recognition · Computer Science 2019-04-04 Sen He , Hamed R. Tavakoli , Ali Borji , Yang Mi , Nicolas Pugeault

Vision-and-language navigation (VLN) aims to enable embodied agents to navigate in realistic environments using natural language instructions. Given the scarcity of domain-specific training data and the high diversity of image and language…

Computer Vision and Pattern Recognition · Computer Science 2021-08-23 Pierre-Louis Guhur , Makarand Tapaswi , Shizhe Chen , Ivan Laptev , Cordelia Schmid

Self-supervised representation learning is able to learn semantically meaningful features; however, much of its recent success relies on multiple crops of an image with very few objects. Instead of learning view-invariant representation…

Computer Vision and Pattern Recognition · Computer Science 2021-10-13 Yuwen Xiong , Mengye Ren , Wenyuan Zeng , Raquel Urtasun

How do humans learn to acquire a powerful, flexible and robust representation of objects? While much of this process remains unknown, it is clear that humans do not require millions of object labels. Excitingly, recent algorithmic…

Computer Vision and Pattern Recognition · Computer Science 2020-10-19 Robert Geirhos , Kantharaju Narayanappa , Benjamin Mitzkus , Matthias Bethge , Felix A. Wichmann , Wieland Brendel

Navigational perception for visually impaired people has been substantially promoted by both classic and deep learning based segmentation methods. In classic visual recognition methods, the segmentation models are mostly object-dependent,…

Computer Vision and Pattern Recognition · Computer Science 2020-07-21 Wei Mao , Jiaming Zhang , Kailun Yang , Rainer Stiefelhagen

This paper presents a comprehensive survey of vision-language (VL) intelligence from the perspective of time. This survey is inspired by the remarkable progress in both computer vision and natural language processing, and recent trends…

Computer Vision and Pattern Recognition · Computer Science 2022-03-04 Feng Li , Hao Zhang , Yi-Fan Zhang , Shilong Liu , Jian Guo , Lionel M. Ni , PengChuan Zhang , Lei Zhang

Pre-training has been adopted in a few of recent works for Vision-and-Language Navigation (VLN). However, previous pre-training methods for VLN either lack the ability to predict future actions or ignore the trajectory contexts, which are…

Computer Vision and Pattern Recognition · Computer Science 2022-03-23 Yanyuan Qiao , Yuankai Qi , Yicong Hong , Zheng Yu , Peng Wang , Qi Wu

Robot navigation with deep reinforcement learning (RL) achieves higher performance and performs well under complex environment. Meanwhile, the interpretation of the decision-making of deep RL models becomes a critical problem for more…

End-to-end training from scratch of current deep architectures for new computer vision problems would require Imagenet-scale datasets, and this is not always possible. In this paper we present a method that is able to take advantage of…

Computer Vision and Pattern Recognition · Computer Science 2017-05-25 Lluis Gomez , Yash Patel , Marçal Rusiñol , Dimosthenis Karatzas , C. V. Jawahar

An emerging paradigm in vision-and-language navigation (VLN) is the use of history-aware multi-modal transformer models. Given a language instruction, these models process observation and navigation history to predict the most appropriate…

Computer Vision and Pattern Recognition · Computer Science 2025-08-14 Dongwoo Kang , Akhil Perincherry , Zachary Coalson , Aiden Gabriel , Stefan Lee , Sanghyun Hong

Navigation is a fundamental capability in embodied AI, representing the intelligence required to perceive and interact within physical environments following language instructions. Despite significant progress in large Vision-Language…

Unlike quasi-static robotic manipulation tasks like pick-and-place, dynamic tasks such as non-prehensile manipulation pose greater challenges, especially for vision-based control. Successful control requires the extraction of features…

In-situ visual observations of marine organisms is crucial to developing behavioural understandings and their relations to their surrounding ecosystem. Typically, these observations are collected via divers, tags, and remotely-operated or…

Computer Vision and Pattern Recognition · Computer Science 2023-05-04 Levi Cai , Nathan E. McGuire , Roger Hanlon , T. Aran Mooney , Yogesh Girdhar