📢 Our paper CubeDiff has been accepted to ICLR 2025! 🎉🎉🎉 CubeDiff achieves SOTA panorama generation with only minimal architectural modifications to existing T2I diffusion models. Check out our website @ and paper @

RT @fedassa: 📣 Google/ETH PhD position available The group of Prof Schindler at ETH offers a fully funded PhD position in collaboration wi…

RT @RdeLutio: Want to remove an undesired object from a scene reconstructed with 3D Gaussian Splatting? We propose an effective solution b…

RT @nicolangnl: Final week to submit your extended abstracts (4-pages) until March 27. All accepted non-archival papers will be presented a…

@Orwelian84 @PeterTor_ @Trolliacc1234 @JanDirkWegner1 Also, our method enables highly efficient pruning, which allows us to run high resolution inference starting at ~11GB VRAM! You can check out Table 2 in the main paper.

@sandrinok @PeterTor_ @JanDirkWegner1 We will soon release it, including trained models!

@songyoupeng Congrats! 🎉👨🏻‍🎓

@nicolangnl @EcoVisionETH @JanDirkWegner1 @walterjetz @blesa_ux @andrescamilor @RdeLutio @GEDI_Knights I like

RT @nicolangnl: Together with @GEDI_knights, we @EcoVisionETH explored the potential of deep learning for interpreting GEDI LIDAR waveforms…