Dylan Campbell

Lecturer @ANUComputing @ANUCECC working on computer vision and machine learning. Previously, postdoc @Oxford_VGG and PhD @CSIRO/@ourANU/NICTA.

[Dylan Campbell @dylanjcampbell_]

We have a new postdoc position at @ANUComputing applying machine learning and optimisation to mechanical engineering (3D shape design). With @sgould_au and @FuaPv, closes 8 March. https://t.co/HwDVLx0qq4

[Stephen Gould @sgould_au]

New postdoc opportunity at @ANUComputing working with myself, @dylanjcampbell_ and @FuaPv on problems related to 3D shape optimization.

Details and applications here:
https://t.co/PHiU9bOZmf

Applications close 8 March 2026.

[Dylan Campbell @dylanjcampbell_]

@shuishidaOx @evan__casey Congratulations Shu! I enjoyed discussing the work with Karl at ICCV - as a former mechanical engineer I love a CAD diagram.

[Dylan Campbell @dylanjcampbell_]

It’s hard to know what’s inside. For point clouds, we can assess this by optimising a winding number field. But the existing formulation treats surface points as infinitely external. We fix this to get faster & better surface reconstruction. Poster #133 this afternoon #ICCV2025

[Chamin Hewa Koneputugodage @ChaminHewa]

📢📢📢 Come check out our poster #𝟏𝟑𝟑: "𝘈𝘯 𝘐𝘮𝘱𝘳𝘰𝘷𝘦𝘥 𝘗𝘰𝘪𝘯𝘵 𝘊𝘭𝘰𝘶𝘥 𝘞𝘪𝘯𝘥𝘪𝘯𝘨 𝘕𝘶𝘮𝘣𝘦𝘳 𝘍𝘰𝘳𝘮𝘶𝘭𝘢𝘵𝘪𝘰𝘯" 𝐭𝐨𝐦𝐨𝐫𝐫𝐨𝐰 (𝐓𝐡𝐮𝐫𝐬𝐝𝐚𝐲 𝐀𝐟𝐭𝐞𝐫𝐧𝐨𝐨𝐧 / 𝐏𝐨𝐬𝐭𝐞𝐫 𝐒𝐞𝐬𝐬𝐢𝐨𝐧 𝟔)

#ICCV2025 @dylanjcampbell_ @sgould_au

[Dylan Campbell @dylanjcampbell_]

@ftm_guney It does seem nonsensical! I've been given advice to not mention conference papers to "avoid giving offense" to other disciplines 😂

[Dylan Campbell @dylanjcampbell_]

Call for papers: Australasian Joint Conference on Artificial Intelligence (#AJCAI2025)
Dates: 1-5 December 2025
Location: Canberra, Australia
Paper Submission: 15 July 2025 AoE
Website: https://t.co/ALhUcKhYRD

A good opportunity to visit Australia!

[Dylan Campbell @dylanjcampbell_]

Call for papers: The 26th International Conference on Digital Image Computing: Techniques and Applications (@dicta2025)
Dates: 3-5 December 2025
Location: Adelaide Convention Centre, Adelaide, Australia
Paper Submission: 15 July, 2025 AoE
Website: https://t.co/sImf3ZQ4vg

[Dylan Campbell @dylanjcampbell_]

@ChuanxiaZ @Oxford_VGG @NTUsg Congratulations Chuanxia! Best wishes for your new position - come visit Australia sometime now that you're closer :)

[Dylan Campbell @dylanjcampbell_]

Congratulations @jianyuan_wang and the VGGT team for winning the #CVPR2025 best paper award - fantastic work!

[Visual Geometry Group (VGG) @Oxford_VGG]

Many Congratulations to @jianyuan_wang, @MinghaoChen23, @n_karaev, Andrea Vedaldi, Christian Rupprecht and @davnov134 for winning the Best Paper Award @CVPR for "VGGT: Visual Geometry Grounded Transformer" 🥇🎉 🙌🙌 #CVPR2025!!!!!!

[Dylan Campbell @dylanjcampbell_]

The Binocular Egocentric 360° workshop is coming to #ICCV2025!

We invite you to participate in the Classification and Temporal Action Localization Kaggle challenges on this unique and uniquely human data (panoramic + binocular egocentric video + audio + binaural delay).

[Machine Intelligence + x (MIx) Group @MIxGroupBham]

🚀 Exciting news! The BinEgo‑360 Workshop&Challenge is coming to #ICCV2025 @ICCVConference! 🌍🎥

We invite you to:
📜 Present your work
🏆 Participate in the Challenge (win a 🌎360 camera!)

💡 360°, ego, multi-modal
🗓️ Challenge DDL: 6th July
🔗 Details: https://t.co/uLA8sUnyZi

[Dylan Campbell @dylanjcampbell_]

A fantastic opportunity with the best supervisors:

[Jakob Foerster @j_foerst]

Joao Henriques (https://t.co/Bedosf2Jti) and I are hiring a fully funded PhD student (UK/international) for the FAIR-Oxford program. The student will spend 50% of their time
@UniofOxford and 50% @AIatMeta (FAIR), while completing a DPhil (Oxford PhD). Deadline: 2nd of Dec AOE!!

[Dylan Campbell @dylanjcampbell_]

Single-view scene reconstruction in a flash!
Augment your favourite monocular depth estimator by predicting multiple Gaussians per pixel, revealing the structure of unseen regions.
Thanks to @StanSzymanowicz, @EldarIsTyping, @ChuanxiaZ, Joao Henriques, @chrirupp & Andrea Vedaldi

[Stan Szymanowicz @StanSzymanowicz]

Feed-forward 3D Gaussians from @Oxford_VGG strike again!
Flash3D has now been accepted to 3DV 2025: it is a method for feed-forward single-view 3D scene reconstruction.

Project page: https://t.co/EcwKdFrLJL
Code: https://t.co/PZ4efKZ3b0
Arxiv: https://t.co/3Nl6UfoJd9

A 🧵👇

[Dylan Campbell @dylanjcampbell_]

Very cool demo from Stan: single image to 3D (Gaussian representation), extremely fast.

[Stan Szymanowicz @StanSzymanowicz]

Announcing The Splatter Image demo🤗
It works on *any* object and is *super* fast - around ~2s per object in Gradio! Trained on 2 GPUs in 3.5 days 🚀
https://t.co/JgU84jSZpW

See 🧵👇 for more results and updates since the original release!

Project: https://t.co/DrwXgm3XII

[Dylan Campbell @dylanjcampbell_]

@ChuanxiaZ @Oxford_VGG @UniofOxford @MSCActions Congratulations Chuanxia!!

[Dylan Campbell @dylanjcampbell_]

@nn0pp0nn @3DVconf We haven't tested it, but I would think so, because it would replace their classification and regression losses during the finetuning stages. Alternatively, their robust formulations could be converted to epipolar equivalents.

[Dylan Campbell @dylanjcampbell_]

SCENES: finetuning correspondence estimators (e.g., LoFTR/Matchformer) with pose-only (or no!) supervision @3DVconf

[Dmytro Mishkin 🇺🇦 @ducha_aiki]

SCENES: Subpixel Correspondence Estimation With Epipolar Supervision

@DominikKloepfer, João Henriques, @dylanjcampbell_

tl;dr: clf + regression epipolar loss for finetuning LoFTR and friends.
https://t.co/DIIvwsrgEE

[Dylan Campbell @dylanjcampbell_]

IMPUS 😈 will be heading to #ICLR2024! Morph between any two images smoothly, directly, and "realistically".

Code coming soon, paper at: https://t.co/9ZRRuAjH3Y

[Frederic Zhang @fredzzhang]

I’ll be presenting our paper in person for the first time! If you are interested in human-object interaction detection, come check out our poster (#147) in Room Nord this morning from 10:30 to 12:30. #ICCV2023 @dylanjcampbell_ @sgould_au

[Dylan Campbell @dylanjcampbell_]

DINO features of the same object vary with viewpoint, making them hard to use for retrieval and scene-level instance segmentation. Visit our poster on Friday morning to learn how to make them view-consistent! #ICCV2023

[Dominik Kloepfer @DominikKloepfer]

📢#ICCV23 is here! I’ll be presenting our paper “LoCUS: Learning Multiscale 3D-Consistent Features from Posed Images” this Friday, 10:30 AM - 12:30 PM in Room Nord.
Swing by, say hi, and let’s chat about cool research!

Details: https://t.co/pZiDRNhDVM 🚀 #ComputerVision

[Dylan Campbell @dylanjcampbell_]

@YicongHong Keep at it Yicong - it's a disheartening process but your talent is formidable, things will go your way soon.

[Dylan Campbell @dylanjcampbell_]

If you aren't able to get to London, AZ's Royal Society lecture will be livestreamed (at 3:30am AEST) and available at: https://t.co/lONqquDLE1. Don't miss it!

[Royal Society Publishing @RSocPublishing]

Join us at the Royal Society this May to hear Professor Andrew Zisserman @Oxford_VGG, winner of the Bakerian Medal and Lecture 2023, deliver his prize lecture on the topic of #ComputerVision. Register now: https://t.co/dl47ghQSUE