Skip to content

Commit ea8e9a0

Browse files
committed
Add new publications, restructure GAIA videos, and fix format
1 parent 2017842 commit ea8e9a0

3 files changed

Lines changed: 154 additions & 59 deletions

File tree

_pages/projects/gaia.md

Lines changed: 65 additions & 46 deletions
Original file line numberDiff line numberDiff line change
@@ -38,24 +38,24 @@ classes: wide
3838
height: 100%;
3939
object-fit: cover;
4040
}
41-
41+
4242
</style>
4343
</head>
4444

4545

4646
<div class="container" style="overflow: hidden;">
47-
<div style="width: 33%; float: left;margin-bottom: 20px; text-align: center;">
47+
<div style="width: 33%; float: left;margin-bottom: 20px; text-align: center;">
4848
<a href="http://www.urjc.es">
49-
<img style="max-width: 60%; height: auto; margin-bottom: 10px;" src="https://upload.wikimedia.org/wikipedia/commons/thumb/8/84/URJC_logo.svg/2560px-URJC_logo.svg.png" alt="">
49+
<img style="max-width: 60%; height: auto; margin-bottom: 10px;" src="https://upload.wikimedia.org/wikipedia/commons/thumb/8/84/URJC_logo.svg/2560px-URJC_logo.svg.png" alt="">
5050
</a>
5151
</div>
52-
<div style="width: 33%; float: left; margin-bottom: 20px; text-align: center;">
52+
<div style="width: 33%; float: left; margin-bottom: 20px; text-align: center;">
5353
<a href="https://roboticslaburjc.github.io/">
54-
<img style="max-width: 30%; height: auto; margin-bottom: 10px;" src="https://pbs.twimg.com/profile_images/1709157413134188544/MHuOqgeK_400x400.jpg" alt="">
54+
<img style="max-width: 30%; height: auto; margin-bottom: 10px;" src="https://pbs.twimg.com/profile_images/1709157413134188544/MHuOqgeK_400x400.jpg" alt="">
5555
</a>
5656
</div>
57-
<div style="width: 33%; float: left; margin-bottom: 100px; text-align: center;">
58-
<img style="max-width: 33%; height: auto; margin-bottom: 10px;" src="/assets/images/lines/gaia.jpg" alt="">
57+
<div style="width: 33%; float: left; margin-bottom: 100px; text-align: center;">
58+
<img style="max-width: 33%; height: auto; margin-bottom: 10px;" src="/assets/images/lines/gaia.jpg" alt="">
5959
</div>
6060
</div>
6161

@@ -69,29 +69,29 @@ Wildfires that burn for weeks and affect millions of people represent a challeng
6969

7070
The main 3 contributions of the RoboticsLab group are:
7171

72-
<b> * LIDAR Signal Densification Algorithms:</b> Research focuses on algorithms and software techniques for densifying the LIDAR sensor signal, commonly used in outdoor robots.
72+
<b> * LIDAR Signal Densification Algorithms:</b> Research focuses on algorithms and software techniques for densifying the LIDAR sensor signal, commonly used in outdoor robots.
7373

74-
<b> * Outdoor Navigation Algorithms:</b> This involves developing and evaluating point-to-point autonomous navigation algorithms for robots in unstructured environments, such as forests.
74+
<b> * Outdoor Navigation Algorithms:</b> This involves developing and evaluating point-to-point autonomous navigation algorithms for robots in unstructured environments, such as forests.
7575

76-
<b> * Deep Learning for Object and Surface Detection:</b> Evolution of techniques for detecting objects and surfaces in unstructured environments using Deep Learning and sensors like LIDAR and visual cameras, aimed at autonomous robot navigation.
76+
<b> * Deep Learning for Object and Surface Detection:</b> Evolution of techniques for detecting objects and surfaces in unstructured environments using Deep Learning and sensors like LIDAR and visual cameras, aimed at autonomous robot navigation.
7777

7878
<div style="display: flex; justify-content: space-around; align-items: flex-start;">
79-
<div style="width: 70%; float: left; margin-bottom: 100px; text-align: center;">
80-
<img style="max-width: 70%; height: auto; margin-bottom: 10px;" src="https://gaia-project.es/media/uploads/bloques/general_/bosque__en_llamas.jpg" alt="">
79+
<div style="width: 70%; float: left; margin-bottom: 100px; text-align: center;">
80+
<img style="max-width: 70%; height: auto; margin-bottom: 10px;" src="https://gaia-project.es/media/uploads/bloques/general_/bosque__en_llamas.jpg" alt="">
8181
</div>
8282
</div>
8383

8484
## Videos
8585

86-
### Lidar Segmentation
86+
### LiDAR Segmentation
8787
<div class="swiper mySwiper">
8888
<div class="swiper-wrapper">
8989
<div class="swiper-slide">
9090
<iframe
91-
id="video-lidar-1"
91+
id="video-lidar-3"
9292
width="560"
9393
height="315"
94-
src="https://www.youtube.com/embed/MCsIj0M7fzM?enablejsapi=1&rel=0&modestbranding=1"
94+
src="https://www.youtube.com/embed/AcJvDmwYneQ?enablejsapi=1&rel=0&modestbranding=1"
9595
title="YouTube video player"
9696
frameborder="0"
9797
allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share"
@@ -100,7 +100,7 @@ The main 3 contributions of the RoboticsLab group are:
100100
</div>
101101
<div class="swiper-slide">
102102
<iframe
103-
id="video-liDAR-2"
103+
id="video-lidar-2"
104104
width="560"
105105
height="315"
106106
src="https://www.youtube.com/embed/qwk5qm3bDOw?enablejsapi=1&rel=0&modestbranding=1"
@@ -112,10 +112,10 @@ The main 3 contributions of the RoboticsLab group are:
112112
</div>
113113
<div class="swiper-slide">
114114
<iframe
115-
id="video-lidar-3"
115+
id="video-lidar-1"
116116
width="560"
117117
height="315"
118-
src="https://www.youtube.com/embed/AcJvDmwYneQ?enablejsapi=1&rel=0&modestbranding=1"
118+
src="https://www.youtube.com/embed/MCsIj0M7fzM?enablejsapi=1&rel=0&modestbranding=1"
119119
title="YouTube video player"
120120
frameborder="0"
121121
allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share"
@@ -128,27 +128,36 @@ The main 3 contributions of the RoboticsLab group are:
128128
<div class="swiper-pagination"></div>
129129
</div>
130130

131-
### Environments and Visual Segmentation
131+
### Image Segmentation
132132
<div class="swiper mySwiper">
133133
<div class="swiper-wrapper">
134134
<div class="swiper-slide">
135135
<iframe
136-
id="video-env-1"
136+
id="video-img-1"
137137
width="560"
138138
height="315"
139-
src="https://www.youtube.com/embed/Zoy12gF2pFw?enablejsapi=1&rel=0&modestbranding=1"
139+
src="https://www.youtube.com/embed/pZQ4GW4Rv3E?enablejsapi=1&rel=0&modestbranding=1"
140140
title="YouTube video player"
141141
frameborder="0"
142142
allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share"
143143
allowfullscreen>
144144
</iframe>
145145
</div>
146+
</div>
147+
<div class="swiper-button-next"></div>
148+
<div class="swiper-button-prev"></div>
149+
<div class="swiper-pagination"></div>
150+
</div>
151+
152+
### Simulated Environments
153+
<div class="swiper mySwiper">
154+
<div class="swiper-wrapper">
146155
<div class="swiper-slide">
147156
<iframe
148-
id="video-env-2"
157+
id="video-env-4"
149158
width="560"
150159
height="315"
151-
src="https://www.youtube.com/embed/ySQJ98mw5pQ?enablejsapi=1&rel=0&modestbranding=1"
160+
src="https://www.youtube.com/embed/W-XWX9uMyDU?enablejsapi=1&rel=0&modestbranding=1"
152161
title="YouTube video player"
153162
frameborder="0"
154163
allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share"
@@ -160,7 +169,7 @@ The main 3 contributions of the RoboticsLab group are:
160169
id="video-env-3"
161170
width="560"
162171
height="315"
163-
src="https://www.youtube.com/embed/ZRoQ_49oX1k?enablejsapi=1&rel=0&modestbranding=1"
172+
src="https://www.youtube.com/embed/Zoy12gF2pFw?enablejsapi=1&rel=0&modestbranding=1"
164173
title="YouTube video player"
165174
frameborder="0"
166175
allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share"
@@ -169,10 +178,10 @@ The main 3 contributions of the RoboticsLab group are:
169178
</div>
170179
<div class="swiper-slide">
171180
<iframe
172-
id="video-env-4"
181+
id="video-env-2"
173182
width="560"
174183
height="315"
175-
src="https://www.youtube.com/embed/W-XWX9uMyDU?enablejsapi=1&rel=0&modestbranding=1"
184+
src="https://www.youtube.com/embed/ySQJ98mw5pQ?enablejsapi=1&rel=0&modestbranding=1"
176185
title="YouTube video player"
177186
frameborder="0"
178187
allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share"
@@ -181,17 +190,17 @@ The main 3 contributions of the RoboticsLab group are:
181190
</div>
182191
<div class="swiper-slide">
183192
<iframe
184-
id="video-env-5"
193+
id="video-env-1"
185194
width="560"
186195
height="315"
187-
src="https://www.youtube.com/embed/pZQ4GW4Rv3E?enablejsapi=1&rel=0&modestbranding=1"
196+
src="https://www.youtube.com/embed/ZRoQ_49oX1k?enablejsapi=1&rel=0&modestbranding=1"
188197
title="YouTube video player"
189198
frameborder="0"
190199
allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share"
191200
allowfullscreen>
192201
</iframe>
193202
</div>
194-
</div>
203+
</div>
195204
<div class="swiper-button-next"></div>
196205
<div class="swiper-button-prev"></div>
197206
<div class="swiper-pagination"></div>
@@ -223,7 +232,7 @@ The main 3 contributions of the RoboticsLab group are:
223232
<div class="swiper-wrapper">
224233
<div class="swiper-slide">
225234
<iframe
226-
id="video-trash-1"
235+
id="video-trash-2"
227236
width="560"
228237
height="315"
229238
src="https://www.youtube.com/embed/E3NFhQJKh64?enablejsapi=1&rel=0&modestbranding=1"
@@ -235,7 +244,7 @@ The main 3 contributions of the RoboticsLab group are:
235244
</div>
236245
<div class="swiper-slide">
237246
<iframe
238-
id="video-trash-2"
247+
id="video-trash-1"
239248
width="560"
240249
height="315"
241250
src="https://www.youtube.com/embed/qjQ6a4S5jp4?enablejsapi=1&rel=0&modestbranding=1"
@@ -295,8 +304,8 @@ The main 3 contributions of the RoboticsLab group are:
295304
},
296305
},
297306
});
298-
299-
307+
308+
300309
// Cargar la API de YouTube
301310
function onYouTubeIframeAPIReady() {
302311
// YouTube IFrames ya estarán listos
@@ -307,36 +316,36 @@ The main 3 contributions of the RoboticsLab group are:
307316
firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);
308317
</script>
309318

310-
<!--
319+
<!--
311320
<div style="display: flex; justify-content: space-around; align-items: flex-start;">
312321
313322
314-
<div style="width: 60%; text-align: center; padding: 10px;">
323+
<div style="width: 60%; text-align: center; padding: 10px;">
315324
<h2>Real-Time 3D Lidar Visualizer</h2>
316325
<iframe width="560" height="315" src="https://www.youtube.com/embed/qwk5qm3bDOw?si=vETtTTi4hGcqeiGt" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>
317326
</div>
318327
319328
320329
321-
<div style="width: 60%; text-align: center; padding: 10px;">
330+
<div style="width: 60%; text-align: center; padding: 10px;">
322331
<h2>Realistic Unstructured environment v3</h2>
323332
<iframe width="560" height="315" src="https://www.youtube.com/embed/Zoy12gF2pFw?si=UcfMKxeauzWipMyf" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>
324333
</div>
325-
326334
327-
335+
336+
328337
329338
</div>
330339
331340
<div style="display: flex; justify-content: space-around; align-items: flex-start;">
332341
333342
334-
<div style="width: 60%; text-align: center; padding: 10px;">
343+
<div style="width: 60%; text-align: center; padding: 10px;">
335344
<h2>Unstructured environment v2</h2>
336345
<iframe width="560" height="315" src="https://www.youtube.com/embed/ySQJ98mw5pQ?si=yR3h9Cu_G6WgOkjS" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>
337346
</div>
338347
339-
<div style="width: 60%; text-align: center; padding: 10px;">
348+
<div style="width: 60%; text-align: center; padding: 10px;">
340349
<h2>Unstructured environment</h2>
341350
<iframe width="560" height="315" src="https://www.youtube.com/embed/ZRoQ_49oX1k?si=eto7gPZoU0TwweIQ" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>
342351
</div>
@@ -347,6 +356,18 @@ The main 3 contributions of the RoboticsLab group are:
347356

348357

349358
## Publications
359+
<div style="display: flex; justify-content: space-around; max-width: 1200px; margin: auto;">
360+
<div style="background-color: #f2f2f2; box-shadow: -8px 0 15px -3px rgba(0, 0, 0, 0.1); padding: 20px; margin: 10px; flex: 1;">
361+
<div style="font-size: 18px; font-weight: bold; margin-bottom: 5px;">Cross-Dataset Evaluation of Visual Semantic Segmentation Models for Off-Road Autonomous Driving</div>
362+
<div style="font-size: 16px; color: #555; margin-bottom: 5px;">David Pascual-Hernández, Sergio Paniego, Roberto Calvo-Palomino, Inmaculada Mora-Jiménez, Jose Maria Cañas-Plaza</div>
363+
<div style="font-size: 14px; color: #555;">Expert Systems with Applications, Elsevier, 2026</div>
364+
</div>
365+
<div style="background-color: #f2f2f2; box-shadow: -8px 0 15px -3px rgba(0, 0, 0, 0.1); padding: 20px; margin: 10px; flex: 1;">
366+
<div style="font-size: 18px; font-weight: bold; margin-bottom: 5px;">Deep Learning-Based Semantic Segmentation of LiDAR Point Clouds for Autonomous Driving in Unstructured Environments</div>
367+
<div style="font-size: 16px; color: #555; margin-bottom: 5px;"> Félix Martínez, David Pascual-Hernández, Daniel Borja Fernández, Inmaculada Mora Jiménez, José María Cañas</div>
368+
<div style="font-size: 14px; color: #555;">Proceedings of the XXV International Workshop on Physical Agents (WAF), 2025</div>
369+
</div>
370+
</div>
350371

351372
<div style="display: flex; justify-content: space-around; max-width: 1200px; margin: auto;">
352373
<div style="background-color: #f2f2f2; box-shadow: -8px 0 15px -3px rgba(0, 0, 0, 0.1); padding: 20px; margin: 10px; flex: 1;">
@@ -374,21 +395,19 @@ The main 3 contributions of the RoboticsLab group are:
374395
<div style="font-size: 14px; color: #555;">SoftwareX, Elsevier, 2024</div>
375396
</div>
376397

377-
398+
378399
</div>
379400

380401

381402

382403

383404
## Founded by
384-
405+
385406

386407

387408
<div class="container" style="overflow: hidden;">
388409
<div style="width: 100%; float: left;margin-bottom: 20px; text-align: center;">
389-
<img style="max-width: 100%; height: auto; margin-bottom: 10px;" src="/assets/images/lines/gaia_founded.png" alt="Paper">
410+
<img style="max-width: 100%; height: auto; margin-bottom: 10px;" src="/assets/images/lines/gaia_founded.png" alt="Paper">
390411
</div>
391-
392-
</div>
393-
394412

413+
</div>
Lines changed: 68 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,68 @@
1+
---
2+
permalink: /publications/2026/cross_dataset_evaluation_of_visual_semantic_segmentation_models_for_off_road_autonomous_driving
3+
title: "Cross-dataset Evaluation of Visual Semantic Segmentation Models for Off-road Autonomous Driving"
4+
5+
6+
layout: single
7+
8+
classes: wide
9+
---
10+
11+
<p style="text-align: center; font-weight: bold;">Neurocomputing, 2024</p>
12+
13+
<p style="text-align: center"><a href="https://sergiopaniego.github.io/">David Pascual-Hernández<sup>1</sup></a>, <a href="https://sergiopaniego.github.io/">Sergio Paniego<sup>1</sup></a>, <a href="https://servicios.urjc.es/pdi/ver/roberto.calvo">Roberto Calvo-Palomino<sup>1</sup></a></p>, <a href="https://servicios.urjc.es/pdi/ver/inmaculada.mora">Inmaculada Mora-Jiménez<sup>1</sup></a></p>, <a href="https://gsyc.urjc.es/jmplaza/">Jose Maria Cañas-Plaza<sup>1</sup></a></p>
14+
<div class="container" style="overflow: hidden;">
15+
<p style="text-align: center; width: 50%; float: left;">1: <a href="https://www.urjc.es/"><img src="https://upload.wikimedia.org/wikipedia/commons/thumb/8/84/URJC_logo.svg/1280px-URJC_logo.svg.png" width="40%" height="40%" alt="URJC"/></a></p>
16+
</div>
17+
<p style="text-align: center">DOI: <a href="https://doi.org/10.1016/j.eswa.2026.132656">10.1016/j.eswa.2026.132656</a></p>
18+
19+
20+
## Abstract
21+
22+
Intelligent autonomous driving in off-road environments is an emerging field with great potential to impact areas such as agriculture, forestry, and rescue operations. Perception in these scenarios presents unique challenges due to the diversity of elements and weather conditions, along with the inherent ambiguity in class definitions. Consequently, off-road visual semantic segmentation datasets remain underdeveloped, roughly ten times smaller than their urban counterparts, hindering dependable performance assessment and potentially compromising the safety of autonomous systems. To address these challenges, we present a comprehensive cross-dataset evaluation of visual semantic segmentation models for autonomous off-road navigation. We propose a unified ontology that harmonizes class definitions across relevant datasets, enabling their combination for both training and testing. This approach ensures fair model comparisons and reliable assessment of generalization to unseen domains. We further benchmark models on the original datasets, analyze the impact of different ontology harmonization criteria and conversion strategies, and evaluate the trade-off between segmentation performance and computational cost. Results show that Transformer-based architectures achieve the most consistent segmentation performance across datasets. While often computationally demanding, some variants maintain real-time inference (≈12 ms) with top-tier accuracy. The unified ontology simplifies the segmentation task, yielding more reliable models and about 40% faster training convergence. Cross-dataset training further enhances generalization, improving mean IoU by up to +20% on RUGD and +13% on WildScenes compared to RELLIS-3D-only training. Overall, this study provides valuable insights for developing robust perception modules for off-road autonomous vehicles.
23+
24+
<div style="display: flex;justify-content: space-around;margin-bottom: 20px;">
25+
<div style="width: 45%;">
26+
<img src="https://ars.els-cdn.com/content/image/1-s2.0-S0957417426015691-gr2_lrg.jpg" frameborder="0" allowfullscreen></img>
27+
</div>
28+
<div style="width: 45%;">
29+
<img src="https://ars.els-cdn.com/content/image/1-s2.0-S0957417426015691-gr11_lrg.jpg" frameborder="0" allowfullscreen></img>
30+
</div>
31+
</div>
32+
33+
## Materials
34+
35+
<div class="container" style="overflow: hidden;">
36+
<div style="width: 33%; float: left;margin-bottom: 20px; text-align: center;">
37+
<a href="https://doi.org/10.1016/j.eswa.2026.132656">
38+
<img style="max-width: 30%; height: auto; margin-bottom: 10px;" src="https://cdn-icons-png.flaticon.com/512/3731/3731553.png" alt="Paper">
39+
<p>Paper</p>
40+
</a>
41+
</div>
42+
<div style="width: 33%; float: left;margin-bottom: 20px; text-align: center;">
43+
<a href="https://github.com/JdeRobot/PerceptionMetrics">
44+
<img style="max-width: 30%; height: auto; margin-bottom: 10px;" src="https://static.thenounproject.com/png/1448954-200.png" alt="Code">
45+
<p>Evaluation Code</p>
46+
</a>
47+
</div>
48+
<div style="width: 33%; float: left;margin-bottom: 20px; text-align: center;">
49+
<a href="https://huggingface.co/roboticslaburjc-org/cross-dataset-visual-offroad-segmentation">
50+
<img style="max-width: 30%; height: auto; margin-bottom: 10px;" src="https://cdn-icons-png.flaticon.com/512/6461/6461819.png" alt="Models' weights">
51+
<p>Models' weights</p>
52+
</a>
53+
</div>
54+
</div>
55+
56+
57+
## Citation
58+
59+
```
60+
@article{pascual2026cross,
61+
title={Cross-Dataset Evaluation of Visual Semantic Segmentation Models for Off-Road Autonomous Driving},
62+
author={Pascual-Hern{\'a}ndez, David and Paniego, Sergio and Calvo-Palomino, Roberto and Mora-Jim{\'e}nez, Inmaculada and Ca{\~n}as-Plaza, Jos{\'e} Mar{\'\i}a},
63+
journal={Expert Systems with Applications},
64+
pages={132656},
65+
year={2026},
66+
publisher={Elsevier}
67+
}
68+
```

0 commit comments

Comments
 (0)