segdac.github.io/index.html at master · SegDAC/segdac.github.io · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
<!DOCTYPE html>
<html>

<head>
  <meta charset="utf-8">
  <!-- Meta tags for social media banners, these should be filled in appropriatly as they are your "business card" -->
  <!-- Replace the content tag with appropriate information -->
  <meta name="description"
    content="SegDAC is a new method for visual reinforcement learning that improves visual generalization by extracting object-centric representations and learning from a dynamic number of segments.">
  <meta property="og:title" content="SegDAC" />
  <meta property="og:description" content="Official project page for SegDAC." />
  <meta property="og:url" content="https://segdac.github.io/" />
  <!-- Path to banner image, should be in the path listed below. Optimal dimenssions are 1200X630-->
  <!-- <meta property="og:image" content="static/image/your_banner_image.png" />
  <meta property="og:image:width" content="1200"/>
  <meta property="og:image:height" content="630"/> -->


  <!-- <meta name="twitter:title" content="TWITTER BANNER TITLE META TAG">
  <meta name="twitter:description" content="TWITTER BANNER DESCRIPTION META TAG">
  Path to banner image, should be in the path listed below. Optimal dimenssions are 1200X600
  <meta name="twitter:image" content="static/images/your_twitter_banner_image.png">
  <meta name="twitter:card" content="summary_large_image"> -->
  <!-- Keywords for your paper to be indexed by-->
  <meta name="keywords"
    content="deep learning,visual reinforcement learning,model-free RL,visual generalization,object-centric">
  <meta name="viewport" content="width=device-width, initial-scale=1">


  <title>SegDAC</title>
  <link rel="icon" type="image/x-icon" href="static/images/favicon.ico">
  <link href="https://fonts.googleapis.com/css?family=Google+Sans|Noto+Sans|Castoro" rel="stylesheet">

  <link rel="stylesheet" href="static/css/bulma.min.css">
  <link rel="stylesheet" href="static/css/bulma-carousel.min.css">
  <link rel="stylesheet" href="static/css/bulma-slider.min.css">
  <link rel="stylesheet" href="static/css/fontawesome.all.min.css">
  <link rel="stylesheet" href="https://cdn.jsdelivr.net/gh/jpswalsh/academicons@1/css/academicons.min.css">
  <link rel="stylesheet" href="static/css/index.css">

  <script src="https://ajax.googleapis.com/ajax/libs/jquery/3.5.1/jquery.min.js"></script>
  <script src="https://documentcloud.adobe.com/view-sdk/main.js"></script>
  <script defer src="static/js/fontawesome.all.min.js"></script>
  <script src="static/js/bulma-carousel.min.js"></script>
  <script src="static/js/bulma-slider.min.js"></script>
  <script src="static/js/index.js"></script>
</head>

<body>


  <section class="hero">
    <div class="hero-body">
      <div class="container is-max-desktop">
        <div class="columns is-centered">
          <div class="column has-text-centered">
            <h1 class="title is-1 publication-title">SegDAC: Visual Generalization in Reinforcement Learning via Dynamic
              Object Tokens</h1>
            <div class="is-size-5 publication-authors">
              <!-- Paper authors -->
              <span class="author-block">
                <a href="https://alexandrebrown.github.io/" target="_blank">Alexandre Brown</a>,</span>
              <span class="author-block">
                <a href="https://neo-x.github.io/" target="_blank">Glen Berseth</a>,</span>
            </div>

            <div class="is-size-5 publication-authors">
              <span class="author-block">Mila Quebec AI Institute, Université de Montréal<br>2025</span>
            </div>

            <div class="column has-text-centered">
              <div class="publication-links">
                <!-- Arxiv PDF link -->
                <span class="link-block">
                  <a href="https://arxiv.org/pdf/2508.09325v4.pdf" target="_blank"
                    class="external-link button is-normal is-rounded is-dark">
                    <span class="icon">
                      <i class="fas fa-file-pdf"></i>
                    </span>
                    <span>Paper</span>
                  </a>
                </span>

                <!-- ArXiv abstract Link -->
                <span class="link-block">
                  <a href="https://arxiv.org/abs/2508.09325" target="_blank"
                    class="external-link button is-normal is-rounded is-dark">
                    <span class="icon">
                      <i class="ai ai-arxiv"></i>
                    </span>
                    <span>arXiv</span>
                  </a>
                </span>

                <!-- Supplementary PDF link
                    <span class="link-block">
                      <a href="static/pdfs/supplementary_material.pdf" target="_blank"
                      class="external-link button is-normal is-rounded is-dark">
                      <span class="icon">
                        <i class="fas fa-file-pdf"></i>
                      </span>
                      <span>Supplementary</span>
                    </a>
                  </span>-->

                <!-- Github link -->
                <span class="link-block">
                  <a href="https://github.com/SegDAC/SegDAC" target="_blank"
                    class="external-link button is-normal is-rounded is-dark">
                    <span class="icon">
                      <i class="fab fa-github"></i>
                    </span>
                    <span>Code</span>
                  </a>
                </span>


              </div>
            </div>
          </div>
        </div>
      </div>
    </div>
  </section>


  <section class="hero teaser">
    <div class="container is-max-desktop">
      <div class="hero-body">
        <video poster="" id="tree" autoplay controls muted loop height="100%">
          <source src="static/videos/segdac.mp4" type="video/mp4">
        </video>
        <h2 class="subtitle has-text-centered">
          SegDAC enables visual RL policies to reason over a dynamic set of object tokens instead of pixels, yielding
          both efficient learning and strong visual generalization.
        </h2>
      </div>
    </div>
  </section>

  <!-- Paper abstract -->
  <section class="section hero is-light">
    <div class="container is-max-desktop">
      <div class="columns is-centered has-text-centered">
        <div class="column is-four-fifths">
          <h2 class="title is-3">Abstract</h2>
          <div class="content has-text-justified">
            <p>
              Visual reinforcement learning policies trained on pixel observations often struggle to generalize when
              visual conditions change at test time. Object-centric representations are a promising alternative, but
              most approaches use fixed-size slot representations, require image reconstruction, or need auxiliary
              losses to learn object decompositions. As a result, it remains unclear how to learn RL policies directly
              from object-level inputs without these constraints. We propose SegDAC, a Segmentation-Driven Actor-Critic
              that operates on a variable-length set of object token embeddings. At each timestep, text-grounded
              segmentation produces object masks from which spatially aware token embeddings are extracted. A
              transformer-based actor-critic processes these dynamic tokens, using segment positional encoding to
              preserve spatial information across objects. We ablate these design choices and show that both segment
              positional encoding and variable-length processing are individually necessary for strong performance. We
              evaluate SegDAC on 8 ManiSkill3 manipulation tasks under 12 visual perturbation types across 3 difficulty
              levels. SegDAC improves over prior visual generalization methods by 15% on easy, 66% on medium, and 88% on
              the hardest settings. SegDAC matches the sample efficiency of the state-of-the-art visual RL methods while
              achieving improved generalization under visual changes.
            </p>
          </div>
        </div>
      </div>
    </div>
  </section>
  <!-- End paper abstract -->


  <!-- Image carousel
<section class="hero is-small">
  <div class="hero-body">
    <div class="container">
      <div id="results-carousel" class="carousel results-carousel">
       <div class="item">
        <img src="static/images/carousel1.jpg" alt="MY ALT TEXT"/>
        <h2 class="subtitle has-text-centered">
          First image description.
        </h2>
      </div>
      <div class="item">
        <img src="static/images/carousel2.jpg" alt="MY ALT TEXT"/>
        <h2 class="subtitle has-text-centered">
          Second image description.
        </h2>
      </div>
      <div class="item">
        <img src="static/images/carousel3.jpg" alt="MY ALT TEXT"/>
        <h2 class="subtitle has-text-centered">
         Third image description.
       </h2>
     </div>
     <div class="item">
      <img src="static/images/carousel4.jpg" alt="MY ALT TEXT"/>
      <h2 class="subtitle has-text-centered">
        Fourth image description.
      </h2>
    </div>
  </div>
</div>
</div>
</section>
-->


  <!-- Youtube video
<section class="hero is-small is-light">
  <div class="hero-body">
    <div class="container">
      <h2 class="title is-3">Video Presentation</h2>
      <div class="columns is-centered has-text-centered">
        <div class="column is-four-fifths">

          <div class="publication-video">
            <iframe src="https://www.youtube.com/embed/JkaxUblCGz0" frameborder="0" allow="autoplay; encrypted-media" allowfullscreen></iframe>
          </div>
        </div>
      </div>
    </div>
  </div>
</section>
-->

  <!--
<section class="hero is-small">
  <div class="hero-body">
    <div class="container">
      <h2 class="title is-3">Another Carousel</h2>
      <div id="results-carousel" class="carousel results-carousel">
        <div class="item item-video1">
          <video poster="" id="video1" autoplay controls muted loop height="100%">

            <source src="static/videos/carousel1.mp4"
            type="video/mp4">
          </video>
        </div>
        <div class="item item-video2">
          <video poster="" id="video2" autoplay controls muted loop height="100%">

            <source src="static/videos/carousel2.mp4"
            type="video/mp4">
          </video>
        </div>
        <div class="item item-video3">
          <video poster="" id="video3" autoplay controls muted loop height="100%">\

            <source src="static/videos/carousel3.mp4"
            type="video/mp4">
          </video>
        </div>
      </div>
    </div>
  </div>
</section> -->


  <!-- Paper poster
<section class="hero is-small is-light">
  <div class="hero-body">
    <div class="container">
      <h2 class="title">Poster</h2>

      <iframe  src="static/pdfs/sample.pdf" width="100%" height="550">
          </iframe>

      </div>
    </div>
  </section>
-->


  <!--BibTex citation -->
  <section class="section" id="BibTeX">
    <div class="container is-max-desktop content">
      <h2 class="title">BibTeX</h2>
      Please use the following BibTeX entry to cite this work:
      <button id="copy-btn" onclick="copyBibtex()" style="
        background-color: #f4f4f5;
        border: 1px solid #d4d4d8;
        border-radius: 6px;
        padding: 6px 12px;
        font-size: 14px;
        cursor: pointer;
        display: flex;
        align-items: center;
        gap: 6px;
        transition: background-color 0.2s ease, border-color 0.2s ease;
      " onmouseover="this.style.backgroundColor='#e4e4e7'; this.style.borderColor='#a1a1aa';"
        onmouseout="this.style.backgroundColor='#f4f4f5'; this.style.borderColor='#d4d4d8';">
        📋 Copy
      </button>
      <pre><code id="bibtex-code">@misc{brown2026segdacvisualgeneralizationreinforcement,
      title={SegDAC: Visual Generalization in Reinforcement Learning via Dynamic Object Tokens},
      author={Alexandre Brown and Glen Berseth},
      year={2026},
      eprint={2508.09325},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2508.09325},
}
</code></pre>
    </div>
  </section>

  <script>
    function copyBibtex() {
      const codeElement = document.getElementById("bibtex-code");
      const btn = document.getElementById("copy-btn");
      const text = codeElement.innerText.trim();

      navigator.clipboard.writeText(text).then(() => {
        btn.textContent = "✅ Copied";
        setTimeout(() => {
          btn.textContent = "📋 Copy";
        }, 1200);
      }).catch(err => {
        console.error("Failed to copy: ", err);
      });
    }
  </script>
  <!--End BibTex citation -->


  <footer class="footer">
    <div class="container">
      <div class="columns is-centered">
        <div class="column is-8">
          <div class="content">

            <p>
              This page was built using the <a href="https://github.com/eliahuhorwitz/Academic-project-page-template"
                target="_blank">Academic Project Page Template</a> which was adopted from the <a
                href="https://nerfies.github.io" target="_blank">Nerfies</a> project page.
              You are free to borrow the source code of this website, we just ask that you link back to this page in the
              footer. <br> This website is licensed under a <a rel="license"
                href="http://creativecommons.org/licenses/by-sa/4.0/" target="_blank">Creative
                Commons Attribution-ShareAlike 4.0 International License</a>.
            </p>

          </div>
        </div>
      </div>
    </div>
  </footer>

  <!-- Statcounter tracking code -->

  <!-- You can add a tracker to track page visits by creating an account at statcounter.com -->

  <!-- End of Statcounter Code -->

</body>

</html>