Merge branch 'master' of github.com:castacks/castacks.github.io

Author: MsFriendly

Gemfile.lock

@@ -78,11 +78,11 @@ GEM
       rb-inotify (~> 0.9, >= 0.9.10)
     mercenary (0.4.0)
     namae (1.1.1)
-    nokogiri (1.18.4-arm64-darwin)
+    nokogiri (1.18.8-arm64-darwin)
       racc (~> 1.4)
-    nokogiri (1.18.4-x86_64-linux-gnu)
+    nokogiri (1.18.8-x86_64-linux-gnu)
       racc (~> 1.4)
-    nokogiri (1.18.4-x86_64-linux-musl)
+    nokogiri (1.18.8-x86_64-linux-musl)
       racc (~> 1.4)
     pathutil (0.16.2)
       forwardable-extended (~> 2.6)

_bibliography/references.bib

@@ -1,3 +1,14 @@
+@misc{alama2025rayfronts,
+      title={RayFronts: Open-Set Semantic Ray Frontiers for Online Scene Understanding and Exploration}, 
+      author={Omar Alama and Avigyan Bhattacharya and Haoyang He and Seungchan Kim and Yuheng Qiu and Wenshan Wang and Cherie Ho and Nikhil Keetha and Sebastian Scherer},
+      year={2025},
+      eprint={2504.06994},
+      archivePrefix={arXiv},
+      primaryClass={cs.RO},
+      url={https://arxiv.org/pdf/2504.06994},
+	  video={https://youtu.be/fFSKUBHx5gA},
+	  abstract={Open-set semantic mapping is crucial for open-world robots. Current mapping approaches either are limited by the depth range or only map beyond-range entities in constrained settings, where overall they fail to combine within-range and beyond-range observations. Furthermore, these methods make a trade-off between fine-grained semantics and efficiency. We introduce RayFronts, a unified representation that enables both dense and beyond-range efficient semantic mapping. RayFronts encodes task-agnostic open-set semantics to both in-range voxels and beyond-range rays encoded at map boundaries, empowering the robot to reduce search volumes significantly and make informed decisions both within & beyond sensory range, while running at 8.84 Hz on an Orin AGX. Benchmarking the within-range semantics shows that RayFronts's fine-grained image encoding provides 1.34x zero-shot 3D semantic segmentation performance while improving throughput by 16.5x. Traditionally, online mapping performance is entangled with other system components, complicating evaluation. We propose a planner-agnostic evaluation framework that captures the utility for online beyond-range search and exploration, and show RayFronts reduces search volume 2.2x more efficiently than the closest online baselines.}
+}
 @misc{baek2025pipe,
       title={PIPE Planner: Pathwise Information Gain with Map Predictions for Indoor Robot Exploration}, 
       author={Seungjae Baek and Brady Moon and Seungchan Kim and Muqing Cao and Cherie Ho and Sebastian Scherer and Jeonghwan Jeon},

_posts/2025-04-10-rayfronts.md

@@ -0,0 +1,23 @@
+---
+layout: post
+title: "RayFronts: Open-Set Semantic Ray Frontiers for Online Scene Understanding and Exploration"
+date: 2025-04-10 10:33:01
+categories: research
+description: "RayFronts is a real-time semantic mapping system that enables fine-grained scene understanding both within and beyond the depth perception range allowing robots to localize and effectively limit search volumes."
+author: "Seungchan Kim"
+published: true
+redirect: "https://rayfronts.github.io/"
+show_sidebar: false
+# slim_content_width: true
+permalink: /rayfronts/
+image: /img/posts/2025-04-10-rayfronts/rayfronts-teaser.gif
+datatable: true
+title_image: None
+hero_image: /img/posts/2025-04-10-rayfronts/rayfronts-teaser.gif
+hero_height: is-large
+remove_hero_title: false
+menubar_toc: false
+tags: Perception, Learning
+---
+
+RayFronts is a real-time semantic mapping system that enables fine-grained scene understanding both within and beyond the depth perception range allowing robots to localize and effectively limit search volumes. RayFronts can be queried with open-set images and text within the map and beyond it. 

_team/adi.md

@@ -1,15 +1,15 @@
 ---
 title: Aditya Rauniyar
-subtitle: Research Associate II
-job_title: Research Associate II
+subtitle: Research Engineer
+job_title: Research Engineer
 category: staff
 layout: team_member_personal_page
 image: /img/team/adi.png
 link-new-tab: true
 ---
 Hi there, this is Adi.
 
-I am a Research Associate at Airlab, working on multi-modal, multi-task learning within multi-agent scenarios. I recently graduated with a Master of Science in Robotics from Carnegie Mellon University, where I focused on 3D computer vision, large-scale training, and reinforcement learning. 
+I am a Research Engineer at Airlab, working on multi-modal, multi-task learning within multi-agent scenarios. I recently graduated with a Master of Science in Robotics from Carnegie Mellon University, where I focused on 3D computer vision, large-scale training, and reinforcement learning. 
 
 [LinkedIn](https://www.linkedin.com/in/adi32/)
 

_team/kabir_kedia.md

@@ -0,0 +1,17 @@
+---
+title: Kabir Kedia
+subtitle: Engineer
+job_title: Engineer
+category: staff
+layout: team_member_personal_page
+image: /img/team/kabir_kedia.jpg
+link-new-tab: true
+---
+Hi there, this is Kabir.
+
+I am an Engineer at Airlab. I recently graduated from Carnegie Mellon University with a Masters in Engineering and AI. 
+
+[LinkedIn](https://www.linkedin.com/in/kabir-kedia/)
+
+[GitHub](https://github.com/kabirkedia)
+