• Skip to primary navigation
  • Skip to main content
SRI logo
  • About
    • Press room
  • Expertise
    • Advanced imaging systems
    • Artificial intelligence
    • Biomedical R&D services
    • Biomedical sciences
    • Computer vision
    • Cyber & formal methods
    • Education and learning
    • Innovation strategy and policy
    • National security
    • Ocean & space
    • Quantum
    • QED-C
    • Robotics, sensors & devices
    • Speech & natural language
    • Video test & measurement
  • Ventures
  • NSIC
  • Careers
  • Contact
  • 日本支社
Search
Close
Home » Conference Paper

Conference Paper

2d-3d reasoning and augmented reality publications May 18, 2022 Conference Paper

Striking the Right Balance: Recall Loss for Semantic Segmentation

Han-Pang Chiu

Class imbalance is a fundamental problem in computer vision applications such as semantic segmentation.

Computer vision publications May 18, 2022 Conference Paper

Graph Mapper: Efficient Visual Navigation by Scene Graph Generation

Han-Pang Chiu, Supun Samarasekera, Rakesh Kumar

We propose a method to train an autonomous agent to learn to accumulate a 3D scene graph representation of its environment by simultaneously learning to navigate through said environment.

Computer vision publications May 18, 2022 Conference Paper

SASRA: Semantically-aware Spatio-temporal Reasoning Agent for Vision-and-Language Navigation in Continuous Environments

Han-Pang Chiu, Supun Samarasekera, Rakesh Kumar

This paper presents a novel approach for the Vision-and-Language Navigation (VLN) task in continuous 3D environments.

2d-3d reasoning and augmented reality publications March 12, 2022 Conference Paper

Head-Worn Markerless Augmented Reality Inside a Moving Vehicle

Han-Pang Chiu, Supun Samarasekera, Rakesh Kumar

This paper describes a system that provides general head-worn outdoor AR capability for the user inside a moving vehicle.

2d-3d reasoning and augmented reality publications January 24, 2022 Conference Paper

SIGNAV: Semantically-Informed GPS-Denied Navigation and Mapping in Visually-Degraded Environments

Han-Pang Chiu, Supun Samarasekera

We present SIGNAV, a real-time semantic SLAM system to operate in perceptually-challenging situations.

Computer vision publications May 30, 2021 Conference Paper

MaAST: Map Attention with Semantic Transformers for Efficient Visual Navigation

Han-Pang Chiu, Supun Samarasekera, Rakesh Kumar

By using our novel attention schema and auxiliary rewards to better utilize scene semantics, we outperform multiple baselines trained with only raw inputs or implicit semantic information while operating with an 80% decrease in the agent’s experience.

Computational sensing-low-power processing publications April 1, 2021 Conference Paper

Hyper-Dimensional Analytics of Video Action at the Tactical Edge

Michael A. Isnardi, David Zhang, Michael Piacentino, Gooitzen van der Wal

We review HyDRATE, a low-SWaP reconfigurable neural network architecture developed under the DARPA AIE HyDDENN (Hyper-Dimensional Data Enabled Neural Network) program. 

Speech & natural language publications July 22, 2020 Conference Paper

Wideband Spectral Monitoring Using Deep Learning

Horacio Franco, Martin Graciarena

We present a system to perform spectral monitoring of a wide band of 666.5 MHz, located within a range of 6 GHz of Radio Frequency (RF) bandwidth, using state-of-the-art deep learning approaches.

2d-3d reasoning and augmented reality publications September 9, 2019 Conference Paper

Semantically-Aware Attentive Neural Embeddings for 2D Long-Term Visual Localization

Han-Pang Chiu, Supun Samarasekera, Rakesh Kumar

We present an approach that combines appearance and semantic information for 2D image-based localization (2D-VL) across large perceptual changes and time lags. Compared to appearance features, the semantic layout of a scene is generally more invariant to appearance variations. We use this intuition and propose a novel end-to-end deep attention-based framework that utilizes multimodal cues to generate robust embeddings for 2D-VL. The proposed attention module predicts a shared channel attention and modality-specific spatial attentions to guide the embeddings to focus on more reliable image regions. We evaluate our model against state-of-the-art (SOTA) methods on three challenging localization datasets. We report an average (absolute) improvement of 19% over current SOTA for 2D-VL. Furthermore, we present an extensive study demonstrating the contribution of each component of our model, showing 8–15% and 4% improvement from adding semantic information and our proposed attention module. We finally show the predicted attention maps to offer useful insights into our model.

  • Go to page 1
  • Go to page 2
  • Go to page 3
  • Interim pages omitted …
  • Go to page 136
  • Go to Next Page »

How can we help?

Once you hit send…

We’ll match your inquiry to the person who can best help you.

Expect a response within 48 hours.

Career call to action image

Make your own mark.

Search jobs

Our work

Case studies

Publications

Timeline of innovation

Areas of expertise

Institute

Leadership

Press room

Media inquiries

Compliance

Careers

Job listings

Contact

SRI Ventures

Our locations

Headquarters

333 Ravenswood Ave
Menlo Park, CA 94025 USA

+1 (650) 859-2000

Subscribe to our newsletter


日本支社
SRI International
  • Contact us
  • Privacy Policy
  • Cookies
  • DMCA
  • Copyright © 2022 SRI International