Recommendation notifications for #3DV2027 are going out by the end of the day (AoE). We’re waiting on a few final meta-reviews.
In case you’re worried about your rebuttal time, author feedback is postponed to November 2 (AoE).
Thanks for your patience!
In case you’re worried about your rebuttal time, author feedback is postponed to November 2 (AoE).
Thanks for your patience!
1
6
32
2,556
🏆 The #3DV2027 Citations Grand Prix finale! P1, with 17208 citations, is “V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation”.
arxiv.org/abs/1606.04797
arxiv.org/abs/1606.04797
1
2
12
1,985
The most cited 3DV paper is a medical one, with more citations than the other four in this countdown combined. V-Net segmented MRI volumes with a 3D fully convolutional network and introduced the Dice loss to handle class imbalance. If you’ve ever used Dice loss, it came from it!
1
1
8
1,502
Nassir Navab’s lab at TUM won both P1 and P3 in the same year at this Grand Prix. Shout-out to Fausto Milletari, @NassirNavab, and @medai_ahmadi!
That’s the podium! Thanks for riding along, and see you in Thessaloniki 🇬🇷
That’s the podium! Thanks for riding along, and see you in Thessaloniki 🇬🇷
0
0
4
311
🌍 3DV Through the Years
Before we reveal P1 of the Citations Grand Prix, a quick detour from papers to places.
3DV started in Seattle in 2013 and is coming to Thessaloniki for #3DV2027. That's a lot of cities in between.
Which 3DV was yours, and what do you remember? 👇
Before we reveal P1 of the Citations Grand Prix, a quick detour from papers to places.
3DV started in Seattle in 2013 and is coming to Thessaloniki for #3DV2027. That's a lot of cities in between.
Which 3DV was yours, and what do you remember? 👇
1
4
16
2,434
🏆 The #3DV2027 Citations Grand Prix continues! P2, with 3337 citations. "Matterport3D: Learning from RGB-D Data in Indoor Environments".
arxiv.org/abs/1709.06158
arxiv.org/abs/1709.06158
1
2
7
1,230
In 2017, RGB-D datasets meant single rooms and hand-held scans. Matterport3D delivered 90 buildings and 194,400 RGB-D images, with camera poses, surface reconstructions, and 2D/3D semantic labels. Indoor scene understanding has been living in these houses ever since.
1
1
2
599
Shout out to @angelxuanchang, @angelaqdai, Thomas Funkhouser, @maciejhalber, @MattNiessner, @ManolisSavva, @SongShuran, @andyzengineer, Yinda Zhang!
🏎️ P1 next. Any guesses?
🏎️ P1 next. Any guesses?
0
0
1
222
🏆 The #3DV2027 Citations Grand Prix continues! P3, with 2793 citations. "Deeper Depth Prediction with Fully Convolutional Residual Networks".
arxiv.org/abs/1606.00373
arxiv.org/abs/1606.00373
1
4
10
1,545
In 2016, depth estimation pipelines typically relied on CRFs or other post-processing. This paper showed that a single fully convolutional residual network with residual upsampling blocks and the reverse Huber loss could outperform them while being smaller and faster. Goated!
1
0
2
542
International Conference on 3D Vision (3DV) retweeted
🍯 Got exciting 3D vision work that deserves another spotlight -- or a bold idea that isn’t quite a paper yet?
Then the #3DV2027 Nectar Track is for you: showcase strong recent work, share early ideas and demos; all while enjoying bougatsa, meze, and rebetiko in beautiful Thessaloniki 🇬🇷
📅 Deadline: Feb 15, 2027 (more details coming soon)
👉 3dvconf.github.io/2027/call-for-…
Then the #3DV2027 Nectar Track is for you: showcase strong recent work, share early ideas and demos; all while enjoying bougatsa, meze, and rebetiko in beautiful Thessaloniki 🇬🇷
📅 Deadline: Feb 15, 2027 (more details coming soon)
👉 3dvconf.github.io/2027/call-for-…
0
12
39
3,187
🏆 The #3DV2027 Citations Grand Prix continues! P4, with 1859 citations. "Towards Linear-time Incremental Structure from Motion" by Changchang Wu!
1
2
12
3,622
The work introduced preemptive feature matching and a BA schedule that make incremental SfM run in linear time in practice, without explicit loop closing. It's the paper behind VisualSFM, the reconstruction tool that a whole generation of 3D vision researchers relied on.
1
0
5
841
A single-author paper from the very first 3DV in 2013. Shout-out to Changchang Wu!
ccwu.me/vsfm/vsfm.pdf
P3 next week 🏎️
ccwu.me/vsfm/vsfm.pdf
P3 next week 🏎️
0
0
2
444
🏆 The #3DV2027 Citations Grand Prix starts now!
Over the next 5 weeks, we'll count down the 5 most-cited papers in 3DV history, from P5 to P1 🏎️
P5, with 1738 citations. "Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervision".
Over the next 5 weeks, we'll count down the 5 most-cited papers in 3DV history, from P5 to P1 🏎️
P5, with 1738 citations. "Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervision".
1
6
17
1,877
The work showed how transfer learning from 2D pose data improves 3D pose estimation in the wild, and it introduced the MPI-INF-3DHP dataset, which remains a standard benchmark in the field. A paper whose influence you can still see in evaluation tables in 2026!
1
0
3
644
Shout out to the team @mehtadushy, @HelgeRhodin, @dancasas, @FuaPv, Oleksandr Sotnychenko, Weipeng Xu, Christian Theobalt!
0
0
4
322
📢The #3DV2027 supplementary material deadline is now September 2 AoE 🌎
Just enough time for your agents to make yet another pass!
📍Thessaloniki, Greece 🇬🇷
3dvconf.github.io/2027
Just enough time for your agents to make yet another pass!
📍Thessaloniki, Greece 🇬🇷
3dvconf.github.io/2027
1
2
19
1,383
📢 The #3DV2027 paper submission deadline is now Aug 28, AoE 🌍
That's about 19 extra hours. Enough time to take a fresh look at some reviews you may have just received and give your 3D work a home it deserves!
📍Thessaloniki, Greece 🇬🇷
3dvconf.github.io/2027
That's about 19 extra hours. Enough time to take a fresh look at some reviews you may have just received and give your 3D work a home it deserves!
📍Thessaloniki, Greece 🇬🇷
3dvconf.github.io/2027
0
3
26
3,176