@francisengelmann.bsky.social
3D Vision and Robotics - PostDoc @ Stanford https://francisengelmann.github.io/
Two more days until #ICCV2025 paper decisions are outπ€If you are working on 3D scene understanding with foundation models and want to present to a thriving community, consider submitting your work to the OpenSUN3D workshop at @iccv.bsky.social ππ
β‘οΈ opensun3d.github.io#paper_track
Join us at OpenSUN3DβοΈ workshop this afternoon
@cvprconference.bsky.social
π
π: Room 105 A
π°οΈ: 2:00-6:00 pm
π: opensun3d.github.io
@lealtaixe.bsky.social @leto--jean.bsky.social
Very excited to announce our final line-up of fantastic speakers at this year's hashtag CVPR workshop on Open-World 3D Scene Understanding with Foundation Models β¨ #OpenSUN3D #cvpr2025 @cvprconference.bsky.social
π June 12, 2pm-6pm
π‘ opensun3d.github.io
What makes a good 3D scene representation? Instead of meshes or Gaussians, we propose Superquadrics to convert scenes into very compact representations β‘οΈ check out our paper for exciting use-cases in roboticsπ€ and GenAIπ
super-dec.github.io
@francisengelmann.bsky.social @marcpollefeys.bsky.social
something is cooking -- πππ
01.06.2025 05:41 β π 12 π 0 π¬ 1 π 1
who says that science doesn't generate profit?
24.04.2025 01:38 β π 51 π 24 π¬ 5 π 7
They even have the golden Athena inside, better than the original π
23.04.2025 04:01 β π 2 π 0 π¬ 0 π 0
5 days left to submit your work on 3D scene understanding with foundation models to the OpenSUN3D workshop @cvprconference.bsky.social
π‘More details: opensun3d.github.io
Deadlines:
π Paper: March 25th
π Challenge: June8th
Could you see how they handle dynamic objects, like moving vehicles, or pedestrians? Seems like parked cars are included in the scans.
21.03.2025 14:02 β π 1 π 0 π¬ 1 π 0
5 days left to submit your work on 3D scene understanding with foundation models to the OpenSUN3D workshop @cvprconference.bsky.social
π‘More details: opensun3d.github.io
Deadlines:
π Paper: March 25th
π Challenge: June8th
That being said, it is actually quite sad that mp4s must be converted to large gifs in 2025 for direct import into google slides.. #Google
18.03.2025 12:49 β π 4 π 1 π¬ 1 π 03D synthetic training data is great, not only for depth -- it provides "perfect" annotations that are difficult to label by hand.
Here we put it to work for human body-part segmentation
human-3d.github.io
πCall for #CVPR2025 workshop papers π
Present your latest work on Open-World 3D Scene Understanding with Foundation ModelsπβοΈ
πPaper Deadline: March 25
(4 pages abstract, or 8 pages paper)
πOpenReview: openreview.net/group?id=thecvf.com/CVPR/2025/Workshop/OpenSUN3D
π‘Details: opensun3d.github.io
I love the new citations preview in overleaf, this is really useful!
π @overleaf.com
π Excited to share our latest work, CrossOver: 3D Scene Cross-Modal Alignment, accepted to #CVPR2025 πβ¨
We learn a unified, modality-agnostic embedding space, enabling seamless scene-level alignment across multiple modalities β no semantic annotations needed!π
π We just added code for quickly visualizing interactions and functionality annotations in 3D scenes.
Find it here: github.com/SceneFun3D/s...
Check out the full #SceneFun3D dataset here:
scenefun3d.github.io
Will also host a #CVPR'25 challenge on this, stay tuned! π
