Georg Martius @ ICML
@gmartius.bsky.social
Prof. Uni Tübingen, Machine Learning, Robotics, Haptics
Here at #EWRL: demonstration of autonomous tomato harvesting by polybot.eu
#Robot #Harvesting #Learning
#Robot #Harvesting #Learning
September 17, 2025 at 1:42 PM
Here at #EWRL: demonstration of autonomous tomato harvesting by polybot.eu
#Robot #Harvesting #Learning
#Robot #Harvesting #Learning
10/10
Surprise bonus 🎁
Without explicit physics priors, 3DGSim also learns lighting & shadows 💡🕶️ as part of “dynamics” — showing its ability to model complex, diverse scene factors 🌍
Surprise bonus 🎁
Without explicit physics priors, 3DGSim also learns lighting & shadows 💡🕶️ as part of “dynamics” — showing its ability to model complex, diverse scene factors 🌍
April 4, 2025 at 9:08 AM
10/10
Surprise bonus 🎁
Without explicit physics priors, 3DGSim also learns lighting & shadows 💡🕶️ as part of “dynamics” — showing its ability to model complex, diverse scene factors 🌍
Surprise bonus 🎁
Without explicit physics priors, 3DGSim also learns lighting & shadows 💡🕶️ as part of “dynamics” — showing its ability to model complex, diverse scene factors 🌍
9/10
Generalization? 🤔
Trained only on single-object-ground collisions, 3DGSim still handles multi-object interactions 📷 — preserving each object’s structure 🧱
Generalization? 🤔
Trained only on single-object-ground collisions, 3DGSim still handles multi-object interactions 📷 — preserving each object’s structure 🧱
April 4, 2025 at 9:08 AM
9/10
Generalization? 🤔
Trained only on single-object-ground collisions, 3DGSim still handles multi-object interactions 📷 — preserving each object’s structure 🧱
Generalization? 🤔
Trained only on single-object-ground collisions, 3DGSim still handles multi-object interactions 📷 — preserving each object’s structure 🧱
7/10
Results!
3DGSim accurately simulates cloth, elastic, and rigid dynamics, capturing realistic motions and interactions across diverse scenarios.
Results!
3DGSim accurately simulates cloth, elastic, and rigid dynamics, capturing realistic motions and interactions across diverse scenarios.
April 4, 2025 at 9:08 AM
7/10
Results!
3DGSim accurately simulates cloth, elastic, and rigid dynamics, capturing realistic motions and interactions across diverse scenarios.
Results!
3DGSim accurately simulates cloth, elastic, and rigid dynamics, capturing realistic motions and interactions across diverse scenarios.
3/10
What’s novel?
3DGSim skips heavy biases (e.g., GNNs) & ground-truth 3D data (e.g., VPD). Training inverse rendering + dynamics end to end lets the encoder learn particle latents that capture both physical and visual aspects.
What’s novel?
3DGSim skips heavy biases (e.g., GNNs) & ground-truth 3D data (e.g., VPD). Training inverse rendering + dynamics end to end lets the encoder learn particle latents that capture both physical and visual aspects.
April 4, 2025 at 9:08 AM
3/10
What’s novel?
3DGSim skips heavy biases (e.g., GNNs) & ground-truth 3D data (e.g., VPD). Training inverse rendering + dynamics end to end lets the encoder learn particle latents that capture both physical and visual aspects.
What’s novel?
3DGSim skips heavy biases (e.g., GNNs) & ground-truth 3D data (e.g., VPD). Training inverse rendering + dynamics end to end lets the encoder learn particle latents that capture both physical and visual aspects.