QMUL Vision Group

Official Twitter account of Computer Vision Group, School of EECS @QMEECS, Queen Mary University of London @QMUL, UK.

[Shitong Sun @sun_shitong315]

Join us at the One Day Meeting: Trustworthy Multimodal Learning with Foundation Models. 🌟 
Details: https://t.co/JOGO2BcCmJ
Location: London EC2R 7BP. 
Let's bridge the gap between AI research and real-world applications! 
#AI #TrustworthyAI #MultimodalLearning

[hu jian @JianHu329]

🔍 Is instance-specific manual prompts a must in SAM? Our AAAI '24 paper segments all the challenging camouflaged objects with only one generic text prompt. No training required! Dive into the details: https://t.co/dW1zekfqMF 🚀 #AI #ComputerVision #AAAI

[British Machine Vision Conference (BMVC) @BMVCconf]

Exciting news! The #BMVC2021 Reviews are now available 🎉🎉. This begins the Revision and Rebuttal Period with the deadline ⏲️Friday 3rd September at 23:59 GMT 👈 remember the timezone!

[QMUL Vision Group @QmulVision]

Your delicate and well-designed deep neutral networks can survive in the low resolution visual recognition challenge? QMUL hold a Low Resolution Image Retrieval challenge https://t.co/b4ZVzQ2V22 in conjunction with the ICCV19’s FORLQ workshop. All welcome! @ICCV19 @CompVisionNews

[HCI Research @HCI_Research]

Now You See It, Now You Don't, Deep Flow-Guided Video Inpainting @cvpr2019 https://t.co/VE8ItCqwQx

[QMUL Vision Group @QmulVision]

Check out our work on unsupervised deep representation learning: "Unsupervised deep learning by neighborhood discovery" on ICML2019! @icmlconf
Paper https://t.co/YLOUmdFuJa
Code:
https://t.co/9vlgEpbP64
We will present it in this Thursday and welcome to Poster#115 for chat!

[Pieter Abbeel @pabbeel]

Looking to study more AI in 2019? All materials of our Fall offering of CS188 Artificial Intelligence are now available here:

https://t.co/M4Mi3UTxyB

@berkeley_ai @UCBerkeley

[Fabio Viola @fabiointheuk]

If you are a PhD student in ML, Neuroscience, CS or a related field, and in the last 2 years of your programme, you should consider applying for an internship at DeepMind! Deadline for next year is Oct 29th - get in touch if you have questions! https://t.co/tYDkvls55O

[Cees Snoek @cgmsnoek]

Looking for two #PhD candidates on data-efficient deep #learning for #video surveillance. Projects in collaboration with #Schiphol and #TNO. Come join the #AI ecosystem in #Amsterdam. https://t.co/mP6MjfV2yT

[Dima Damen @dimadamen]

@VisionInstitute @QmulVision glad to be hosting Eddy Zhu to talk about scalable person Re-ID @UoB_Engineering

[Chen Change Loy @ccloy]

mmdetection is released! https://t.co/2b2Vs8ZrOz
- Modular design
- Supports popular detection frameworks, e.g. Faster RCNN, Mask RCNN, RetinaNet, etc.
- The training speed is ~5%-20% faster than Detectron for different models.
- Codebase of COCO object detection winning team

[InnvtUK-Smart-City-Challenge @iUK_Smart_City]

Check out our new open logo detection challenge: https://t.co/wra2LrWE9h, appeared n BMVC 2018.

[Visual Geometry Group (VGG) @Oxford_VGG]

OxUVA - The largest dataset for training/testing long-term object tracking! (presented at #ECCV18, VGG+TVG+QUVA collaboration) #acronyms #objecttracking #dataset
https://t.co/nPVnZJ1GXQ

[Chen Change Loy @ccloy]

Call for paper: #IJCV Special Issue on Generative Adversarial Networks for Computer Vision. Full paper submission deadline: 31 Mar 2019

[🌲Roelof ☀️ @graphific]

One of the most comprehensive overviews of the past decade of breakthroughs in object recognition (going over architectures, context modelling, detection proposal methods, datasets, and evaluation criteria) #DeepLearning https://t.co/siHiiyHEev

[Chen Change Loy @ccloy]

Stay tuned, we are going to release the codebase of the MMDet team, who won the #ECCV2018 #COCO Detection 2018 Challenge. https://t.co/2b2Vs8ZrOz

[QMUL Vision Group @QmulVision]

#ECCV2018 P-4B-63: SketchyScene: Richly-Annotated Scene Sketches.

[QMUL Vision Group @QmulVision]

#ECCV2018 We have two papers to be presented on 12 Sep:

(1) P-3A-24 Pose-Normalized Image Generation for Person Re-identification

(2) P-3C-49 Deep Factorised Inverse-Sketching

[QMUL Vision Group @QmulVision]

Two papers are presented today (11 Sep) at #ECCV2018:

(1) P-2A-01Unsupervised Person Re-identification by Deep Learning Tracklet Association, Li et al.;

(2) P-2B-19Universal Sketch Perceptual Grouping, Li et al..