Maximilian Denninger (@DenningerMax) header
Maximilian Denninger (@DenningerMax) profile picture

Maximilian Denninger

@DenningerMax

Deep learning engineer at Apple. Former PhD cand. at German Aerospace Center (DLR). All opinions are my own.

 München, Deutschland
Joined December 2016
  • Tweets82
  • Following253
  • Followers304
4Media
Photo by Maximilian Denninger (@DenningerMax)Photo by Maximilian Denninger (@DenningerMax)Photo by Maximilian Denninger (@DenningerMax)Photo by Maximilian Denninger (@DenningerMax)
Maximilian Denninger retweeted
Markus Knauer (@KnauerMarkus) profile picture
Markus Knauer@KnauerMarkusMar 14
🚀 New paper in IEEE RA-L! IROSA: Interactive Robot Skill Adaptation using Natural Language Tell a robot "go faster" or "there's an obstacle" - it adapts. The LLM selects validated tool functions, no direct robot control: a zero-shot, local LLM. #robotics #LLM #IROS2026 #HRI 👇
1
1
3
466
Maximilian Denninger retweeted
Markus Knauer (@KnauerMarkus) profile picture
Markus Knauer@KnauerMarkus19 May 2025
Check out our paper presented at #ICRA2025 How can robotic systems articulate truthful answers about their internal knowledge in natural language? 🤖 📅 Presentation: May 20th, 2025, at 16:45 EDT in Room A314 📍 Poster session: Immediately following the talks
Image from Markus Knauer (@KnauerMarkus)
Image from Markus Knauer (@KnauerMarkus)
1
1
2
91
Maximilian Denninger retweeted
Markus Knauer (@KnauerMarkus) profile picture
Markus Knauer@KnauerMarkus7 Mar 2025
🚀 Exciting news! Our work has been published in IEEE 𝘙𝘰𝘣𝘰𝘵𝘪𝘤𝘴 𝘢𝘯𝘥 𝘈𝘶𝘵𝘰𝘮𝘢𝘵𝘪𝘰𝘯 𝘓𝘦𝘵𝘵𝘦𝘳𝘴 (𝘙𝘈-𝘓). See links in comments.
2
1
2
93
Starting my dream job at Apple today! I am so excited working on new and challenging problems.
1
15
BlenderProc 2.5.0 is out! Features: - segmentations now in the same rendering call - support for MatterPort3d - a Kinect depth noise model - intersecting spotlights with the camera frustum, creating hard lighting situations And much more! github.com/DLR-RM/BlenderPro…
7
24
Maximilian Denninger retweeted
Martin Sundermeyer (@ma_sundermeyer) profile picture
Martin Sundermeyer@ma_sundermeyer22 Sep 2022
Submit your results to the #BOP Challenge in the next 4 weeks and show the world that your method works outside the LineMod/YCBV playground! Awards are golden this year 😉
Eric Brachmann (@eric_brachmann) profile pictureEric Brachmann@eric_brachmann22 Sep 2022
We are happy to announce that the winners of the #BOP challenge 2022 will be awarded the total of $4000! The prize is sponsored by Niantic and Meta who recognise the key role that robust and precise 6D pose estimation will play in #AR/#VR and #robotics in the years to come.
Image from Eric Brachmann (@eric_brachmann)
1
1
11
BlenderProc 2.4.0 is out! It now supports URDF files, so you can load and manipulate your own robot! github.com/DLR-RM/BlenderPro…
GitHub - DLR-RM/BlenderProc: A procedural Blender pipeline for photorealistic training image...

GitHub - DLR-RM/BlenderProc: A procedural Blender pipeline for photorealistic training image...

A procedural Blender pipeline for photorealistic training image generation - DLR-RM/BlenderProc

github.com
1
11
41
Maximilian Denninger retweeted
Eric Brachmann (@eric_brachmann) profile picture
Eric Brachmann@eric_brachmann12 Jul 2022
Got a 6D pose paper accepted to #CVPR2022 or #ECCV2022? Great! You convinced the reviewers, now convince everyone else. The #BOP challenge accepts your submission until 14th Oct. bop.felk.cvut.cz/challenges/… @ma_sundermeyer @tomhodan @eccvconf Your paper was rejected? 🧵👇
Image from Eric Brachmann (@eric_brachmann)
1
6
36
Maximilian Denninger retweeted
Martin Sundermeyer (@ma_sundermeyer) profile picture
Martin Sundermeyer@ma_sundermeyer8 Jun 2022
The Benchmark on Object Pose Estimation 2022 is open! What is the actual best method today? bop.felk.cvut.cz/challenges/… Submit your results until 14.10.22 📅 Win awards at our R6D workshop @ECCV22 🏆 cmp.felk.cvut.cz/sixd/worksh… #CVPR #BOP #sim2real @tomhodan
Image from Martin Sundermeyer (@ma_sundermeyer)
3
7
2
31
Maximilian Denninger retweeted
Martin Sundermeyer (@ma_sundermeyer) profile picture
Martin Sundermeyer@ma_sundermeyer11 Mar 2022
Our new work presents a very efficient 6-DoF object tracker & refiner. Looking forward to a real conference @ #CVPR2022 :)
Manuel (@StoiberMa) profile pictureManuel@StoiberMa11 Mar 2022
Happy to share our work "Iterative Corresponding Geometry (ICG)" accepted to #CVPR2022! SOTA 6DoF tracking, requiring only the object geometry and 1.3ms per frame on a single CPU core. arXiv: arxiv.org/abs/2203.05334 Code: github.com/DLR-RM/3DObjectTr… Video: youtu.be/qMr1RHCsnDk
Iterative Corresponding Geometry: Fusing Region and Depth for...

Iterative Corresponding Geometry: Fusing Region and Depth for...

Tracking objects in 3D space and predicting their 6DoF pose is an essential task in computer vision. State-of-the-art approaches often rely on object texture to tackle this problem. However, while...

arxiv.org
3
12
BlenderProc 2.2.0 now supports blender 3.0.0! This means improved render times, making the generation of the perfect dataset even faster. Go check it out yourself: github.com/DLR-RM/BlenderPro… #MachineLearning #datasets #blender3d
GitHub - DLR-RM/BlenderProc: A procedural Blender pipeline for photorealistic training image...

GitHub - DLR-RM/BlenderProc: A procedural Blender pipeline for photorealistic training image...

A procedural Blender pipeline for photorealistic training image generation - DLR-RM/BlenderProc

github.com
1
35
3
191
For a full list of changes see our change log: github.com/DLR-RM/BlenderPro…
1
3
BlenderProc got to a 1.000 stars on GitHub today. I can’t believe we got so far, when I started this years ago, I never imagined that so many people would use BlenderProc! github.com/DLR-RM/BlenderPro…
GitHub - DLR-RM/BlenderProc: A procedural Blender pipeline for photorealistic training image...

GitHub - DLR-RM/BlenderProc: A procedural Blender pipeline for photorealistic training image...

A procedural Blender pipeline for photorealistic training image generation - DLR-RM/BlenderProc

github.com
3
11
52
Maximilian Denninger retweeted
Klaus H. Strobl (@5trobl) profile picture
Klaus H. Strobl@5trobl17 Nov 2021
Contributing to #BlenderProc you end up contributing to yourself. The tidy, well-organized code, their programming rules, and first and foremost the supportive and fun authors made this contribution my favorite task of the year @DenningerMax @ma_sundermeyer @cornerfarmer
Maximilian Denninger (@DenningerMax) profile pictureMaximilian Denninger@DenningerMax17 Nov 2021
Check out what @5trobl helped us to achieve:
1
10
Check out what @5trobl helped us to achieve:
Maximilian Denninger (@DenningerMax) profile pictureMaximilian Denninger@DenningerMax17 Nov 2021
BlenderProc 2.1.0 is out! Blender excels in photometric realism w.r.t. actual images; now BlenderProc 2.1.0 adds pixel-accurate geometric realism as measured by geometric camera calibration. github.com/DLR-RM/BlenderPro… #blender #blenderproc #deeplearning #machinelearning
1
5
BlenderProc 2.1.0 is out! Blender excels in photometric realism w.r.t. actual images; now BlenderProc 2.1.0 adds pixel-accurate geometric realism as measured by geometric camera calibration. github.com/DLR-RM/BlenderPro… #blender #blenderproc #deeplearning #machinelearning
1
12
3
43
We have a nice example to show it is done: dlr-rm.github.io/BlenderProc…
1
1
PS: Check we did even more, for full details check out our change log: github.com/DLR-RM/BlenderPro…
1
BlenderProc2 is out! So many new features a full python API and an easy command line interface. Explore it now, with a simple command: pip install blenderproc github.com/DLR-RM/BlenderPro… #blenderproc #blender #deeplearning #machinelearning
1
34
7
175
Load more