Oliver Batchelor
@oliverbatch.bsky.social
Computer Vision/Agritech researcher designing a vineyard/orchard many-camera scanner (Gaussian Splatting-based).
Radiance fields, 3D recognition, FP fan (Haskell).
Running, rock climbing, dog friends.
ucvision.org - University of Canterbury, New Zealand
Radiance fields, 3D recognition, FP fan (Haskell).
Running, rock climbing, dog friends.
ucvision.org - University of Canterbury, New Zealand
Pinned
Oliver Batchelor
@oliverbatch.bsky.social
· Nov 20
both robots
YouTube video by Oliver Batchelor
youtu.be
This is what we do! We scan plants with multiple cameras/other sensors. Extract 3d models then extract structure and plant organs. These are a couple of our older scanners.
Reposted by Oliver Batchelor
What AI person is saying.....
Anthropic CEO, Dario Amodei: In the next 3 to 6 months, AI is writing 90% of the code, and in 12 months, nearly all code may be generated by AI.
What AI person is doing.....
www.anthropic.com/jobs
Anthropic CEO, Dario Amodei: In the next 3 to 6 months, AI is writing 90% of the code, and in 12 months, nearly all code may be generated by AI.
What AI person is doing.....
www.anthropic.com/jobs
March 12, 2025 at 8:48 PM
What AI person is saying.....
Anthropic CEO, Dario Amodei: In the next 3 to 6 months, AI is writing 90% of the code, and in 12 months, nearly all code may be generated by AI.
What AI person is doing.....
www.anthropic.com/jobs
Anthropic CEO, Dario Amodei: In the next 3 to 6 months, AI is writing 90% of the code, and in 12 months, nearly all code may be generated by AI.
What AI person is doing.....
www.anthropic.com/jobs
Reposted by Oliver Batchelor
We recently released the code for "Efficient Perspective-Correct 3D Gaussian Splatting Using Hybrid Transparency"
Project Page: fhahlbohm.github.io/htgs/
Code: github.com/nerficg-proj...
Project Page: fhahlbohm.github.io/htgs/
Code: github.com/nerficg-proj...
March 8, 2025 at 11:57 AM
We recently released the code for "Efficient Perspective-Correct 3D Gaussian Splatting Using Hybrid Transparency"
Project Page: fhahlbohm.github.io/htgs/
Code: github.com/nerficg-proj...
Project Page: fhahlbohm.github.io/htgs/
Code: github.com/nerficg-proj...
Reposted by Oliver Batchelor
February 5, 2025 at 6:09 PM
Reposted by Oliver Batchelor
#3D imaging data collected on French vines from Champagne allows us to precisely study the internal degradation of vine trunks. 🍇🩻🧲
Using #X-rays and #MRI #3D imaging
👉 Find out more at rdcu.be/dAOuD
@IFV @CIRAD @ComitéChampagne @INRAE @AGAP @PlanDepVignoble
#UMT_genovigne #trunk_diseases #IA
Using #X-rays and #MRI #3D imaging
👉 Find out more at rdcu.be/dAOuD
@IFV @CIRAD @ComitéChampagne @INRAE @AGAP @PlanDepVignoble
#UMT_genovigne #trunk_diseases #IA
November 18, 2024 at 9:44 AM
#3D imaging data collected on French vines from Champagne allows us to precisely study the internal degradation of vine trunks. 🍇🩻🧲
Using #X-rays and #MRI #3D imaging
👉 Find out more at rdcu.be/dAOuD
@IFV @CIRAD @ComitéChampagne @INRAE @AGAP @PlanDepVignoble
#UMT_genovigne #trunk_diseases #IA
Using #X-rays and #MRI #3D imaging
👉 Find out more at rdcu.be/dAOuD
@IFV @CIRAD @ComitéChampagne @INRAE @AGAP @PlanDepVignoble
#UMT_genovigne #trunk_diseases #IA
Reposted by Oliver Batchelor
🌿New non-destructive approach:
🧲☢️
#MRI & #XRAY CT scans combined with #MachineLearning reveal grapevine trunk tissue degradation at different stages, achieving 91% accuracy. 🏆✅
Full paper 👉 www.nature.com/articles/s41...
#PlantHealth #Grapevine 🍇 #imaging #3D #AI
#IFV @cirad.bsky.social #Champagne
🧲☢️
#MRI & #XRAY CT scans combined with #MachineLearning reveal grapevine trunk tissue degradation at different stages, achieving 91% accuracy. 🏆✅
Full paper 👉 www.nature.com/articles/s41...
#PlantHealth #Grapevine 🍇 #imaging #3D #AI
#IFV @cirad.bsky.social #Champagne
December 4, 2024 at 2:37 PM
🌿New non-destructive approach:
🧲☢️
#MRI & #XRAY CT scans combined with #MachineLearning reveal grapevine trunk tissue degradation at different stages, achieving 91% accuracy. 🏆✅
Full paper 👉 www.nature.com/articles/s41...
#PlantHealth #Grapevine 🍇 #imaging #3D #AI
#IFV @cirad.bsky.social #Champagne
🧲☢️
#MRI & #XRAY CT scans combined with #MachineLearning reveal grapevine trunk tissue degradation at different stages, achieving 91% accuracy. 🏆✅
Full paper 👉 www.nature.com/articles/s41...
#PlantHealth #Grapevine 🍇 #imaging #3D #AI
#IFV @cirad.bsky.social #Champagne
Reposted by Oliver Batchelor
I was born in socialism, but grew up in capitalism. I saw the shift from books of my childhood - idolizing scientists, engineers, doctors, adventurers, martyrs and quiet self-sacrifice, working towards the common good - to pop-culture of money, success, status, being loud, and bling. 3/
December 27, 2024 at 3:43 PM
I was born in socialism, but grew up in capitalism. I saw the shift from books of my childhood - idolizing scientists, engineers, doctors, adventurers, martyrs and quiet self-sacrifice, working towards the common good - to pop-culture of money, success, status, being loud, and bling. 3/
Reposted by Oliver Batchelor
“Once men turned their thinking over to machines in the hope that this would set them free. But that only permitted other men with machines to enslave them.”
― Frank Herbert, Dune
― Frank Herbert, Dune
December 8, 2024 at 5:36 PM
“Once men turned their thinking over to machines in the hope that this would set them free. But that only permitted other men with machines to enslave them.”
― Frank Herbert, Dune
― Frank Herbert, Dune
Reposted by Oliver Batchelor
I stumbled on @eugenevinitsky.bsky.social 's blog and his "Personal Rules of Productive Research" is very good. I now do a lot of things in the post, & wish I had done them when I was younger.
I share my "mini-paper" w ppl I hope will be co-authors.
www.eugenevinitsky.com/posts/person...
I share my "mini-paper" w ppl I hope will be co-authors.
www.eugenevinitsky.com/posts/person...
Eugene Vinitsky
www.eugenevinitsky.com
December 16, 2024 at 3:14 PM
I stumbled on @eugenevinitsky.bsky.social 's blog and his "Personal Rules of Productive Research" is very good. I now do a lot of things in the post, & wish I had done them when I was younger.
I share my "mini-paper" w ppl I hope will be co-authors.
www.eugenevinitsky.com/posts/person...
I share my "mini-paper" w ppl I hope will be co-authors.
www.eugenevinitsky.com/posts/person...
Reposted by Oliver Batchelor
I am very excited to announce our paper “DynaMoN: Motion-Aware Fast and Robust Camera Localization for Dynamic Neural Radiance Fields” has been accepted to #IEEE RA-L.
Paper: ieeexplore.ieee.org/document/107...
Project page: hannahhaensen.github.io/DynaMoN/
Code: github.com/HannahHaense...
Paper: ieeexplore.ieee.org/document/107...
Project page: hannahhaensen.github.io/DynaMoN/
Code: github.com/HannahHaense...
December 9, 2024 at 11:39 AM
I am very excited to announce our paper “DynaMoN: Motion-Aware Fast and Robust Camera Localization for Dynamic Neural Radiance Fields” has been accepted to #IEEE RA-L.
Paper: ieeexplore.ieee.org/document/107...
Project page: hannahhaensen.github.io/DynaMoN/
Code: github.com/HannahHaense...
Paper: ieeexplore.ieee.org/document/107...
Project page: hannahhaensen.github.io/DynaMoN/
Code: github.com/HannahHaense...
Reposted by Oliver Batchelor
To kick off using Bluesky: our new dataset called Oxford Spires. Synchronised, multi-color cameras and lidar in multiple Oxford colleges. Ground-Truth highly accurate 3D maps from tripod scanners. The ideal basis for NeRF/3DGS SLAM research.
dynamic.robots.ox.ac.uk/datasets/oxf...
dynamic.robots.ox.ac.uk/datasets/oxf...
December 9, 2024 at 10:29 AM
To kick off using Bluesky: our new dataset called Oxford Spires. Synchronised, multi-color cameras and lidar in multiple Oxford colleges. Ground-Truth highly accurate 3D maps from tripod scanners. The ideal basis for NeRF/3DGS SLAM research.
dynamic.robots.ox.ac.uk/datasets/oxf...
dynamic.robots.ox.ac.uk/datasets/oxf...
Reposted by Oliver Batchelor
Welcome to all the newcomers! Here's a starter pack for the 3D Computer Vision community: go.bsky.app/Cfm9XFe
November 28, 2024 at 9:30 AM
Welcome to all the newcomers! Here's a starter pack for the 3D Computer Vision community: go.bsky.app/Cfm9XFe
Reposted by Oliver Batchelor
Adding my love letter to
arxiv.org/pdf/2304.01315
Empirical Design in Reinforcement Learning
by
Andrew Patterson, Samuel Neumann, Martha White, Adam White
JMLR 25 (2024) 1-63
#ReinforcementLearning
These aren’t the heroes we deserve, but they are the heroes we need.
arxiv.org/pdf/2304.01315
Empirical Design in Reinforcement Learning
by
Andrew Patterson, Samuel Neumann, Martha White, Adam White
JMLR 25 (2024) 1-63
#ReinforcementLearning
These aren’t the heroes we deserve, but they are the heroes we need.
arxiv.org
November 23, 2024 at 1:40 PM
Adding my love letter to
arxiv.org/pdf/2304.01315
Empirical Design in Reinforcement Learning
by
Andrew Patterson, Samuel Neumann, Martha White, Adam White
JMLR 25 (2024) 1-63
#ReinforcementLearning
These aren’t the heroes we deserve, but they are the heroes we need.
arxiv.org/pdf/2304.01315
Empirical Design in Reinforcement Learning
by
Andrew Patterson, Samuel Neumann, Martha White, Adam White
JMLR 25 (2024) 1-63
#ReinforcementLearning
These aren’t the heroes we deserve, but they are the heroes we need.
It seems like a torch.compile-based MLP has now caught up with tiny-cuda-nn (at least for the workloads I use).. and is more flexible.
Has anyone experimented with putting LayerNorm in their NeRF? In some experiments I'm doing it seems to make things a lot more stable.
Has anyone experimented with putting LayerNorm in their NeRF? In some experiments I'm doing it seems to make things a lot more stable.
November 23, 2024 at 5:11 AM
It seems like a torch.compile-based MLP has now caught up with tiny-cuda-nn (at least for the workloads I use).. and is more flexible.
Has anyone experimented with putting LayerNorm in their NeRF? In some experiments I'm doing it seems to make things a lot more stable.
Has anyone experimented with putting LayerNorm in their NeRF? In some experiments I'm doing it seems to make things a lot more stable.
Reposted by Oliver Batchelor
After my general computer vision starter pack is now full (150/150 entries reached), here is one specific to 3D Vision: go.bsky.app/Cfm9XFe
November 21, 2024 at 8:15 AM
After my general computer vision starter pack is now full (150/150 entries reached), here is one specific to 3D Vision: go.bsky.app/Cfm9XFe
The radio on the way home tonight… “Mr. Blue Sky, please tell us why
You had to hide away for so long (so long)
Where did we go wrong?”
Seemed symbolic!
You had to hide away for so long (so long)
Where did we go wrong?”
Seemed symbolic!
November 20, 2024 at 11:00 AM
The radio on the way home tonight… “Mr. Blue Sky, please tell us why
You had to hide away for so long (so long)
Where did we go wrong?”
Seemed symbolic!
You had to hide away for so long (so long)
Where did we go wrong?”
Seemed symbolic!
Reposted by Oliver Batchelor
November 19, 2024 at 11:00 PM
This is what we do! We scan plants with multiple cameras/other sensors. Extract 3d models then extract structure and plant organs. These are a couple of our older scanners.
both robots
YouTube video by Oliver Batchelor
youtu.be
November 20, 2024 at 10:16 AM
This is what we do! We scan plants with multiple cameras/other sensors. Extract 3d models then extract structure and plant organs. These are a couple of our older scanners.
Reposted by Oliver Batchelor
Here's my attempt at assembling a starter pack of the still nascent computer vision community on Bluesky. Feel free to recommend other accounts that should be in this starter pack. :) #computervision (Do we use hashtags here? 😅)
go.bsky.app/PkAKJu5
go.bsky.app/PkAKJu5
November 11, 2024 at 8:08 AM
Here's my attempt at assembling a starter pack of the still nascent computer vision community on Bluesky. Feel free to recommend other accounts that should be in this starter pack. :) #computervision (Do we use hashtags here? 😅)
go.bsky.app/PkAKJu5
go.bsky.app/PkAKJu5
Reposted by Oliver Batchelor
Bring back academic Twitter! Maybe I'll start posting papers on here too :)
November 13, 2024 at 2:13 PM
Bring back academic Twitter! Maybe I'll start posting papers on here too :)
Reposted by Oliver Batchelor
One thing still missing here is good discussion of academic papers, so I guess I'll be the change I want to see in the world
Really interesting results from Jacob Andreas's group showing great performance on ARC-AGI just from doing a few gradient descent steps at test time arxiv.org/abs/2411.07279
Really interesting results from Jacob Andreas's group showing great performance on ARC-AGI just from doing a few gradient descent steps at test time arxiv.org/abs/2411.07279
The Surprising Effectiveness of Test-Time Training for Abstract Reasoning
Language models have shown impressive performance on tasks within their training distribution, but often struggle with novel problems requiring complex reasoning. We investigate the effectiveness of t...
arxiv.org
November 13, 2024 at 1:06 PM
One thing still missing here is good discussion of academic papers, so I guess I'll be the change I want to see in the world
Really interesting results from Jacob Andreas's group showing great performance on ARC-AGI just from doing a few gradient descent steps at test time arxiv.org/abs/2411.07279
Really interesting results from Jacob Andreas's group showing great performance on ARC-AGI just from doing a few gradient descent steps at test time arxiv.org/abs/2411.07279
arxiv.org/abs/2411.06390
"SplatFormer takes as input an initial 3DGS set optimized under limited training views and refines it in a single forward pass, effectively removing potential artifacts in OOD test views"
Impressive, fixing novel viewpoints - but scenes seem to be from their dataset.
"SplatFormer takes as input an initial 3DGS set optimized under limited training views and refines it in a single forward pass, effectively removing potential artifacts in OOD test views"
Impressive, fixing novel viewpoints - but scenes seem to be from their dataset.
November 14, 2024 at 1:35 AM
arxiv.org/abs/2411.06390
"SplatFormer takes as input an initial 3DGS set optimized under limited training views and refines it in a single forward pass, effectively removing potential artifacts in OOD test views"
Impressive, fixing novel viewpoints - but scenes seem to be from their dataset.
"SplatFormer takes as input an initial 3DGS set optimized under limited training views and refines it in a single forward pass, effectively removing potential artifacts in OOD test views"
Impressive, fixing novel viewpoints - but scenes seem to be from their dataset.
This place may not yet have a thriving community on topics I'm interested in. I'm done with being a second-class citizen on a platform which promotes right-wing shit-stirring.
November 13, 2024 at 11:22 AM
This place may not yet have a thriving community on topics I'm interested in. I'm done with being a second-class citizen on a platform which promotes right-wing shit-stirring.
Where are the machine learning and computer vision people here? I want to quit Twitter, but I can't find the same topics.
November 11, 2024 at 6:32 AM
Where are the machine learning and computer vision people here? I want to quit Twitter, but I can't find the same topics.
Is it just me or is bsky.app hard to remember? I have to go to google and search for "bluesky" to find it every time.
Perhaps I just have poor memory.
Perhaps I just have poor memory.
Bluesky
Social media as it should be. Find your community among millions of users, unleash your creativity, and have some fun again.
bsky.app
September 11, 2024 at 12:32 PM
Is it just me or is bsky.app hard to remember? I have to go to google and search for "bluesky" to find it every time.
Perhaps I just have poor memory.
Perhaps I just have poor memory.