Skip to content

Commit 2979b8d

Browse files
SimeonOASimeonOA
committed
Cleared everything till I am ready to update
1 parent ab95507 commit 2979b8d

File tree

1 file changed

+26
-26
lines changed

1 file changed

+26
-26
lines changed

‎index.html

Lines changed: 26 additions & 26 deletions
Original file line numberDiff line numberDiff line change
@@ -166,7 +166,7 @@ <h2 class="subheader">Under Review</h1>
166166
<div class="section base-row add-top-padding">
167167
<h1 class="tldr">
168168
<b>TL;DR</b>:
169-
Robot See Robot Do uses a <b>4D</b> <b>D</b>ifferentiable <b>P</b>art <b>M</b>odel (4D-DPM) to visually imitate articulated motions from an object scan and single monocular video.
169+
Botany-Bot uses Gaussian Splatting and Garfield to create a segmented 3D reconstruction of plants and then uses this to guide a robot's interaction with the plant to reveal even more information.
170170
</h1>
171171
<video id="main-video" autobuffer muted autoplay loop controls playsinline>
172172
<source id="mp4" src="data/supp_video.mp4" type="video/mp4">
@@ -181,7 +181,7 @@ <h1>Abstract</h1>
181181
</p>
182182
</div>
183183

184-
<div class="section base-row add-top-padding">
184+
<!-- <div class="section base-row add-top-padding">
185185
<h1 class="tldr">
186186
Humans imitate manipulation by watching <b>object</b> motion, not <b>hand</b> motion. RSRD does the same. <br>
187187
This enables imitation from a single video robust to orientation and across the human-robot embodiment gap.
@@ -195,9 +195,9 @@ <h1 class="tldr">
195195
<img src="data/favicon.png" style="height: 40px; margin-left: 5px; margin-right: 5px; margin-top: 40px; margin-bottom: 40px;">
196196
<img src="data/favicon.png" style="height: 40px; margin-left: 5px; margin-right: 5px; margin-top: 40px; margin-bottom: 40px;">
197197
</div>
198-
</div>
198+
</div> -->
199199

200-
<div class="section base-row add-top-padding">
200+
<!-- <div class="section base-row add-top-padding">
201201
202202
203203
<h1>4D Part Reconstruction</h1>
@@ -206,15 +206,15 @@ <h1>4D Part Reconstruction</h1>
206206
By creating part-aware 3D representations using <a href="https://www.garfield.studio/" class="author-text" target="_blank">GARField</a> (parts, toggle for clusters) and
207207
<a href="https://github.com/facebookresearch/dinov2" class="author-text"target="_blank">DINOv2</a> (tracking SE3 pose),
208208
these smartphone-captured inputs can generate these 4D reconstructions:
209-
</p>
209+
</p> -->
210210

211211

212-
<div id="main-results">
213-
<h1 class="tldr"><b>Input demonstration video</b></h1>
214-
<div class="video-carousel-wrapper">
215-
<div class="carousel-container" id="videoCarousel">
212+
<!-- <div id="main-results">
213+
<h1 class="tldr"><b>Input demonstration video</b></h1> -->
214+
<!-- <div class="video-carousel-wrapper">
215+
<div class="carousel-container" id="videoCarousel"> -->
216216
<!-- Add your video elements here -->
217-
<div class="carousel-item" id="bear_video">
217+
<!-- <div class="carousel-item" id="bear_video">
218218
<video autoplay muted loop playsinline height="400px">
219219
<source src="data/demo_vids/cal_bear_naked_wave.mp4" type="video/mp4">
220220
</video>
@@ -233,8 +233,8 @@ <h1 class="tldr"><b>Input demonstration video</b></h1>
233233
<video autoplay muted loop playsinline height="400px">
234234
<source src="data/demo_vids/scissors2.mp4" type="video/mp4">
235235
</video>
236-
</div>
237-
<div class="carousel-item" id="sunglasses_video">
236+
</div> -->
237+
<!-- <div class="carousel-item" id="sunglasses_video">
238238
<video autoplay muted loop playsinline height="400px">
239239
<source src="data/demo_vids/sunglasses_fold.mp4" type="video/mp4">
240240
</video>
@@ -260,12 +260,12 @@ <h1 class="tldr"><b>Input demonstration video</b></h1>
260260
</video>
261261
</div>
262262
</div>
263-
</div>
263+
</div> -->
264264
<!-- <h1 class="tldr"><b>
265265
<img src="data/drag_icon.png" alt="" class="inline-image">Click and move me!<img src="data/drag_icon.png" alt="" class="inline-image">
266266
</b></h1> -->
267267

268-
<div id="iframe-container" class="iframe-container">
268+
<!-- <div id="iframe-container" class="iframe-container">
269269
<div class="click-and-move-overlay">
270270
<h1 class="tldr">
271271
<b>
@@ -351,9 +351,9 @@ <h1 class="tldr">
351351
<img src="data/favicon.png" style="height: 40px; margin-left: 5px; margin-right: 5px; margin-top: 40px; margin-bottom: 40px;">
352352
<img src="data/favicon.png" style="height: 40px; margin-left: 5px; margin-right: 5px; margin-top: 40px; margin-bottom: 40px;">
353353
</div>
354-
</div>
354+
</div> -->
355355

356-
<div class="section base-row add-top-padding">
356+
<!-- <div class="section base-row add-top-padding">
357357
<h1>Robot Motion Retargeting</h1>
358358
<p class="paragraph">
359359
After recovering 3D part motion, RSRD optimizes grasps and robot motions to reproduce the 4D reconstructions.
@@ -464,10 +464,10 @@ <h1 class="tldr">
464464
</div>
465465
</div>
466466
</p>
467-
</div>
467+
</div> -->
468468

469469
<!-- Just for vertical spacing... -->
470-
<div class="section add-top-padding">
470+
<!-- <div class="section add-top-padding">
471471
<div class="base-row">
472472
<img src="data/favicon.png" style="height: 40px; margin-left: 5px; margin-right: 5px; margin-top: 40px; margin-bottom: 40px;">
473473
<img src="data/favicon.png" style="height: 40px; margin-left: 5px; margin-right: 5px; margin-top: 40px; margin-bottom: 40px;">
@@ -489,9 +489,9 @@ <h1>How it works</h1>
489489
Because 4D-DPM uses gradient descent, any differentiable prior is easily incorporated like temporal smoothness and rigidity.
490490
</p>
491491
<img src="data/4ddpm_ablation.jpg" style="max-width: 90%" />
492-
</div>
492+
</div> -->
493493

494-
<div class="section base-row add-top-padding">
494+
<!-- <div class="section base-row add-top-padding">
495495
<h1>Retargeting Robot Trajectory</h1>
496496
<p class="paragraph">
497497
With the recovered 3D part motion and the object placed in the robot workspace,
@@ -556,14 +556,14 @@ <h1>Retargeting Robot Trajectory</h1>
556556
</p>
557557
</div>
558558
<!-- Just for vertical spacing... -->
559-
<div class="section add-top-padding">
559+
<!-- <div class="section add-top-padding">
560560
<div class="base-row">
561561
<img src="data/favicon.png" style="height: 40px; margin-left: 5px; margin-right: 5px; margin-top: 40px; margin-bottom: 40px;">
562562
<img src="data/favicon.png" style="height: 40px; margin-left: 5px; margin-right: 5px; margin-top: 40px; margin-bottom: 40px;">
563563
<img src="data/favicon.png" style="height: 40px; margin-left: 5px; margin-right: 5px; margin-top: 40px; margin-bottom: 40px;">
564564
</div>
565-
</div>
566-
<div class="section base-row add-top-padding">
565+
</div> --> -->
566+
<!-- <div class="section base-row add-top-padding">
567567
<h1>Limitations and Failures</h1>
568568
<p class="paragraph">
569569
4D monocular reconstruction is an extremely under-constrained and challenging problem, and 4D-DPM still suffers from sensitivity to demonstration viewpoint, occlusions, reconstruction quality, and more.
@@ -631,10 +631,10 @@ <h1>Limitations and Failures</h1>
631631
</video>
632632
</div>
633633
</div>
634-
</div>
634+
</div> -->
635635

636636

637-
<div class="section citation" style="margin-top: 50px">
637+
<!-- <div class="section citation" style="margin-top: 50px">
638638
<h1 id="abstract">Citation </h1>
639639
<p class="paragraph"> If you use this work or find it helpful, please consider citing: (bibtex) </p>
640640
<pre id="codecell0">@inproceedings{kerr2024rsrd,
@@ -643,7 +643,7 @@ <h1 id="abstract">Citation </h1>
643643
&nbsp;booktitle={8th Annual Conference on Robot Learning},
644644
&nbsp;year={2024},
645645
&nbsp;url={https://openreview.net/forum?id=2LLu3gavF1}
646-
}
646+
} -->
647647
</pre>
648648
</div>
649649
<footer>

0 commit comments

Comments
 (0)