Hu, Cheng, Yu et al., "VGGT4D: Mining Motion Cues in Visual Geometry Transformers for 4D Scene Reconstruction"
Easi3r-style attention analysis and masking with mask refinement with VGGT. Also discards tokens related to dynamic points.
Personal programs for ICCV 2025 are now available at:
www.scholar-inbox.com/conference/i...
Look, 4D foundation models know about humans โ and we just read it out!
Glad to be recognized as an outstanding reviewer!
๐Page: rover-xingyu.github.io/TTT3R
๐Paper: arxiv.org/abs/2509.26645
๐ปCode: github.com/Inception3D/...
Big thanks to the amazing team!
@xingyu-chen.bsky.social @fanegg.bsky.social @xiuyuliang.bsky.social @andreasgeiger.bsky.social @apchen.bsky.social
Instead of updating all states uniformly, we incorporate image attention as per-token learning rates.
High-confidence matches get larger updates, while low-quality updates are suppressed.
This soft gating greatly extends the length generalization beyond the training context.
#VGGT: accurate within short clips, but slow and prone to Out-of-Memory (OOM)
#CUT3R: fast with constant memory usage, but forgets.
We revisit them from a Test-Time Training (TTT) perspective and propose #TTT3R to get all three: fast, accurate, and OOM-free.
Let's keep revisiting 3D reconstruction!
Excited to introduce LoftUp!
A strong (than ever) and lightweight feature upsampler for vision encoders that can boost performance on dense prediction tasks by 20%โ100%!
Easy to plug into models like DINOv2, CLIP, SigLIP โ simple design, big gains. Try it out!
github.com/andrehuang/l...
If you're a researcher and haven't tried it yet, please give it a try! It took me a while to adjust, but now it's my favorite tool. You can read, bookmark, organize papers, and get recommendations based on your interests!
Easi3R: Estimating Disentangled Motion from DUSt3R Without Training
@xingyu-chen.bsky.social, @fanegg.bsky.social, @xiuyuliang.bsky.social, @andreasgeiger.bsky.social, @apchen.bsky.social
arxiv.org/abs/2503.24391
๐๐ฎ๐๐ถ๐ฏ๐ฅ: ๐๐๐๐ถ๐บ๐ฎ๐๐ถ๐ป๐ด ๐๐ถ๐๐ฒ๐ป๐๐ฎ๐ป๐ด๐น๐ฒ๐ฑ ๐ ๐ผ๐๐ถ๐ผ๐ป ๐ณ๐ฟ๐ผ๐บ ๐๐จ๐ฆ๐๐ฏ๐ฅ ๐ช๐ถ๐๐ต๐ผ๐๐ ๐ง๐ฟ๐ฎ๐ถ๐ป๐ถ๐ป๐ด
Xingyu Chen, Yue Chen, Yuliang Xiu ... Anpei Chen
arxiv.org/abs/2503.24391
Trending on www.scholar-inbox.com
DUSt3R was never trained to do dynamic segmentation with GT masks, right? It was just trained to regress point maps on 3D datasetsโyet dynamic awareness emerged, making DUSt3R a zero-shot 4D estimator!๐
I was really surprised when I saw this. Dust3R has learned very well to segment objects without supervision. This knowledge can be extracted post-hoc, enabling accurate 4D reconstruction instantly.
๐Page: easi3r.github.io
๐Paper: arxiv.org/abs/2503.24391
๐ปCode: github.com/Inception3D/...
Big thanks to the amazing team!
@xingyu-chen.bsky.social, @fanegg.bsky.social, @xiuyuliang.bsky.social, @andreasgeiger.bsky.social, @apchen.bsky.social
With our estimated segmentation masks, we perform a second inference pass by re-weighting the attention, enabling robust 4D reconstruction and even outperforming SOTA methods trained on 4D datasets, with almost no extra cost compared to vanilla DUSt3R.
We propose an attention-guided strategy to decompose dynamic objects from the static background, enabling robust dynamic object segmentation. It outperforms the optical-flow guided segmentation, like MonST3R, and the model trained on dynamic mask labels, like DAS3R.
๐กHumans naturally separate ego-motion from object-motion without dynamic labels. We observe that #DUSt3R has implicitly learned a similar mechanism, reflected in its attention layers.
๐ฆฃEasi3R: 4D Reconstruction Without Training!
Limited 4D datasets? Take it easy.
#Easi3R adapts #DUSt3R for 4D reconstruction by disentangling and repurposing its attention maps โ make 4D reconstruction easier than ever!
๐Page: easi3r.github.io
How much 3D do visual foundation models (VFMs) know?
Previous work requires 3D data for probing โ expensive to collect!
#Feat2GS @cvprconference.bsky.social 2025 - our idea is to read out 3D Gaussains from VFMs features, thus probe 3D with novel view synthesis.
๐Page: fanegg.github.io/Feat2GS
