Skip to content

Commit 27c4d9d

Browse files
DAG-DiffDAG-Diff
authored
Update index.html
1 parent 27926e3 commit 27c4d9d

1 file changed

Lines changed: 45 additions & 18 deletions

File tree

‎index.html‎

Lines changed: 45 additions & 18 deletions
Original file line numberDiff line numberDiff line change
@@ -49,32 +49,54 @@ <h1 class="title is-1 publication-title"><span class="gradient-text">DAGDiff</sp
4949
Dual-Arm Grasp Diffusion to Stable and Collision-Free Grasps
5050
</h1>
5151

52+
<div style="
53+
display: inline-block;
54+
padding: 2px;
55+
border-radius: 12px;
56+
background: linear-gradient(90deg, hsl(251, 100%, 65%), hsl(278, 100%, 73%));
57+
margin-bottom: 15px;
58+
">
59+
<div style="
60+
background: white;
61+
border-radius: 10px;
62+
padding: 12px 18px;
63+
text-align: center;
64+
">
65+
<h2 class="title is-5" style="margin: 0;">
66+
Accepted at ICRA 2026
67+
</h2>
68+
</div>
69+
</div>
70+
71+
5272
<div class="is-size-5 publication-authors">
5373
<span class="author-block">
5474
<a href="https://researchweb.iiit.ac.in/~md.faizal/" target="_blank">Md Faizal
5575
Karim</a><sup>1⭐</sup>,</span>
5676
<span class="author-block">
57-
<a href="https://in.linkedin.com/in/vignesh-vembar" target="_blank">Vignesh Vembar</a><sup>1⭐</sup>,</span>
77+
<a href="https://in.linkedin.com/in/vignesh-vembar" target="_blank">Vignesh
78+
Vembar</a><sup>1⭐</sup>,</span>
5879
<span class="author-block">
59-
<a href="https://sites.google.com/site/keshabpatra19" target="_blank">Keshab Patra</a><sup>1</sup>,
80+
<a href="https://sites.google.com/site/keshabpatra19" target="_blank">Keshab
81+
Patra</a><sup>1</sup>,
6082
</span>
6183
<span class="author-block">
6284
<a href="https://vanhalen42.github.io/" target="_blank">Gaurav Singh</a><sup>2</sup>,
6385
</span>
6486
<span class="author-block">
65-
<a href="https://scholar.google.com/citations?user=QDuPGHwAAAAJ" target="_blank">K Madhava Krishna</a><sup>1</sup>
87+
<a href="https://scholar.google.com/citations?user=QDuPGHwAAAAJ" target="_blank">K
88+
Madhava Krishna</a><sup>1</sup>
6689
</span>
67-
</div>
68-
90+
</div>
91+
6992
<div class="is-size-5 publication-authors">
7093
<span class="author-block"><sup>1</sup>IIIT Hyderabad </span>
7194
<span class="author-block"><sup>2</sup>Brown University</span>
7295
<br>
7396
<span class="author-block"><sup>⭐</sup>Equal contribution</span>
7497
<br>
7598
</div>
76-
77-
<!--
99+
78100

79101
<div class="column has-text-centered">
80102
<div class="publication-links">
@@ -88,14 +110,15 @@ <h1 class="title is-1 publication-title"><span class="gradient-text">DAGDiff</sp
88110
</a>
89111
</span>
90112

91-
<!-- <span class="link-block">
92-
<a href="#" class="external-link button is-normal is-rounded is-dark">
113+
<span class="link-block">
114+
<a href="https://arxiv.org/abs/2509.21145"
115+
class="external-link button is-normal is-rounded is-dark">
93116
<span class="icon">
94117
<i class="ai ai-arxiv"></i>
95118
</span>
96119
<span>arXiv</span>
97120
</a>
98-
</span> -->
121+
</span>
99122
<span class="link-block">
100123
<a href="#video" class="external-link button is-normal is-rounded is-dark">
101124
<span class="icon">
@@ -105,7 +128,7 @@ <h1 class="title is-1 publication-title"><span class="gradient-text">DAGDiff</sp
105128
</a>
106129
</span>
107130
<span class="link-block">
108-
<a href="https://anonymous.4open.science/r/dual-arm-grasp-diffusion-E5AF/README.md"
131+
<a href="https://github.com/DAG-Diff/dual-arm-grasp-diffusion"
109132
class="external-link button is-normal is-rounded is-dark">
110133
<span class="icon">
111134
<i class="fab fa-github"></i>
@@ -506,25 +529,29 @@ <h3 class="title is-5 mt-5 pt-5">3. Real-World Dual-Arm Grasp Results</h3>
506529
</div>
507530

508531
<div class="content has-text-justified my-4">
509-
<b>Quantitative Results:</b> DAGDiff consistently outperforms across all evaluation settings. It outperforms prior
510-
methods in <u>Force-Closure Evaluation (FCE)</u> and <u>Grasp Success Rate (GSR)</u> while maintaining the lowest <u>Grasp
511-
Collision Rate (GCR)</u>, indicating more physically valid and robust dual-arm grasps. In zero-shot transfer
532+
<b>Quantitative Results:</b> DAGDiff consistently outperforms across all evaluation settings. It
533+
outperforms prior
534+
methods in <u>Force-Closure Evaluation (FCE)</u> and <u>Grasp Success Rate (GSR)</u> while maintaining
535+
the lowest <u>Grasp
536+
Collision Rate (GCR)</u>, indicating more physically valid and robust dual-arm grasps. In zero-shot
537+
transfer
512538
to Dual-Afford objects, DAGDiff continues to show strong generalization without task-specific
513539
retraining. Finally, real-world experiments on unseen objects such as trays, buckets, and pans
514540
demonstrate consistent success, confirming that DAGDiff’s classifier-guided diffusion produces grasps
515-
that are stable, collision-free, and transferable beyond simulation. Real-life failures occur mostly due to noisy point-cloud estimation and
541+
that are stable, collision-free, and transferable beyond simulation. Real-life failures occur mostly due
542+
to noisy point-cloud estimation and
516543
hance generated grasps in these cases are not always perfect.
517544
</div>
518545
</div>
519546
</section>
520547

521548

522-
<section class="section" id="BibTeX" style="margin-bottom: 1rem;">
549+
<!-- <section class="section" id="BibTeX" style="margin-bottom: 1rem;">
523550
<div class="container is-max-desktop content">
524551
<h2 class="title">BibTeX</h2>
525552
<pre><code>Will be updated</code></pre>
526553
</div>
527-
</section>
554+
</section> -->
528555

529556
<section class="section" id="BibTeX" style="margin-bottom: 1rem;">
530557
<div class="container is-max-desktop content">
@@ -548,7 +575,7 @@ <h2 class="title">BibTeX</h2>
548575
This website is licensed under a <a rel="license"
549576
href="http://creativecommons.org/licenses/by-sa/4.0/">Creative
550577
Commons Attribution-Share Alike 4.0 International License</a>. It is borrowing this <a
551-
href="https://github.com/nerfies/nerfies.github.io">source code</a>
578+
href="https://github.com/nerfies/nerfies.github.io">source code.</a>
552579
</p>
553580
</div>
554581
</div>

0 commit comments

Comments
 (0)