Group Meetings

Schedule 2019

Tentative schedule:
June 6: Mikel
May 30:Lu
May 9: Lichao
May 2: Fei and Kai (his own)
Apr 25: Marc

Apr 9: Marc and Lu present their own work.

Apr 4: Kai wil present
Hardness-Aware Deep Metric Learning (CVPR2019 oral)

Oguz presents his work.

Mar 28: Lichao will present his work.

Carola will present:
Large Scale Fine-Grained Categorization and Domain-Specific Transfer
(CVPR 2018)

Mar 7: Chenshen will present:
The lottery ticket hypothesis: finding sparse, trainable neural networks
(ICLR 2019 Oral)

Feb 28: Xialei will present:
On the Sensitivity of Adversarial Robustness to Input Data Distributions (ICLR 2019)

Oguz will present:
Semantic Regularisation for Recurrent Image Annotation

Feb 14: Yaxing will present:
Unsupervised Learning of Object Landmarks through Conditional Image Generation

Javad will present:
Large-Scale Visual Active Learning with Deep Probabilistic Ensembles


Dec 12 Lichao will present his recent work.

Kai will present Robust Classification with Convolutional Prototype Learning (CVPR2018)

Nov 28 Lu will present the paper: Learning to Compare: Relation Network for Few-Shot Learning.

Yaxing will present his recent research.

Nov 7: Carola will present Associating Inter-image Salient Instances for Weakly Supervised Semantic Segmentation (ECCV 2018)

Bogdan will present Progress & Compress: A scalable framework for continual learning (ICLR 2018)

Oct 31: Chenshen will present Large Scale GAN Training for High Fidelity Natural Image Synthesis

Oct 17: Oguz will present Deep Randomized Ensembles for Metric Learning (ECCV 2018)

Abel will present What do I Annotate Next? An Empirical Study of Active Learning for Action Localization

Oct 10: Marc presents Memory Aware Synapses: Learning what (not) to forget

Sept 19: Fei presents DF-Net: Unsupervised Joint Learning of Depth and Flow using Cross-Task Consistency (ECCV 2018)

Laura presents: Group Normalization (ECCV 2018)

Sept 5:Xialei presents Deep Matching Autoencoders (CVPR 2018)

13- 6 Kai will present Self-supervised learning of visual features through embedding images into text topic spaces (CVPR2017)

Chenshen will present his latest research.

30-5 CVPR Meeting: Lu will present VITON: An Image-based Virtual Try-on Network

Chenshen will present Unsupervised Feature Learning via Non-Parametric Instance Discrimination

Fei will present Context-aware Synthesis for Video Frame Interpolation

Laura will present Look, Imagine and Match: Improving Textual-Visual Cross-Modal Retrieval With Generative Models

25-5 Profesor Henry Medeiros gives a talk in the group: Applications of Semantic Image Segmentation in Agricultural Automation and Assisted Living Environments

23-5 Carola will present: Weakly Supervised Salient Object Detection Using Image Labels

Oguz will present: Learning Attribute Representations With Localization for Flexible Fashion Search

Lichao will present: Fast and Accurate Online Video Object Segmentation via Tracking Parts

11-5 Lu will discuss here submission to BMVC.

2-5 Luis will present ‘FiLM: Visual Reasoning with a General Conditioning Layer’ and Abel will present ‘Multimodal Unsupervised Image-to-Image Translation‘.

18-4 CVC CVPR meeting (Abel, Xialei, Yaxing, Javier)

5-4 Yaxing presented his submission to ECCV.

21-3 ICLR seminars: Talk 1: Skip RNN: Learning to Skip State Updates in Recurrent Neural Networks, by Victor Campos, Talk 2: Graph Attention Networks, by Guillem Cucurull

7-3 Xialei presents Learning to Segment Every Thing

21-2 Marc presents BIER – Boosting Independent Embeddings Robustly


7-2 Javad presents Apprentice: Using Knowledge Distillation Techniques To Improve Low-Precision Network Accuracy

31-1 Bogdan presents Photographic Image Synthesis with Cascaded Refinement Networks

25-1 Aitor presents ‘Learning to Synthesize a 4D RGBD Light Field from a Single Image‘ (ICCV)

17-1 Abel represents his work and we will discuss: Unsupervised Creation of Parameterized Avatars

12-1 Jan van Gemert presents a seminar.


20-12 Luis will present ‘progressive growing of GANs’

15-12 Abel will present: Interpretable and Globally Optimal Prediction for Textual Grounding using Image Concepts

and Lu will present: Memory-Augmented Attribute Manipulation Networks for Interactive Fashion Search

5-12 Fei will present Optical Flow with Semantic Segmentation and Localized Layers

30-11 Chenshen presents NIPS 2017: ‘Continual Learning with Deep Generative Replay’

and Kai will be present ACM-MM ‘Fluency-Guided Cross-Lingual Image Captioning’

24-11 Xialei and Yaxing present their work.

7-11 Deeper, Broader and Artier Domain Generalization

18-10 Presentation from Marco Buzzelli on his thesis research sofar.

11-10 Xialei on learning without forgetting

6-10 Marc on New Pattern Discovery

26-9 CVPR/ICCV meeting:
Focal Loss for Dense Object Detection, ICCV 2017 (Javad)
Top-down Visual Saliency Guided by Captions, CVPR 2017 (Carola)

20-9 CVPR/ICCV meeting:
Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization ,ICCV2017 (Yaxing)
Adversarial Discriminative Domain Adaptation, CVPR 2017 (Oğuz)

15-9 CVPR/ICCV meeting:
Incremental Learning of Object Detectors without Catastrophic Forgetting, ICCV2017, (Xialei)
FC4: Fully Convolutional Color Constancy With Confidence-Weighted Pooling, CVPR 2017 (Rada)

15-9 Jan van Gemert is visiting and gives a seminar on ‘Active Decision Boundary Annotation with Deep Generative Models’

6-9 CVPR/ICCV meeting: CREST: Convolutional Residual Learning for Visual Tracking.(Lichao) Attend in groups: a weakly-supervised deep learning framework for learning from web data.(Lu) On Compressing Deep Models by Low Rank and Sparse Decomposition (Marc)

1-9 Abel Gonzalez presents his thesis work.

31-8 We will discuss the paper ‘YOLO9000: Better, Faster, Stronger’

22-7 Oguz presents his work

17-7 The CVC CVPR reading group.

10-7 We discuss the paper Learning Aligned Cross-Modal Representations from Weakly Aligned Data

29-6 Carola presents her work

22-6 Lichao presents his work, and we will discuss the paper ‘Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks’

8-6 Rada and Lu presents their work

25-5 Xialei presents his work

18-5 We will discuss the paper BEGAN: Boundary Equilibrium Generative Adversarial Networks

Marc presents his work

11-5 Yaxing presents his work on ‘Multiresolution GANs’

28-4 ICLR meeting3 (short presentations):

20-4 ICLR meeting2 (short presentations):
ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation
Highway and Residual Networks learn Unrolled Iterative Estimation

18-4 We have a talk from Damien Muselet ‘Towards an Electronic Orientation Table: Using Features Extracted From the Image to Register Digital Elevation Model’

6-4 ICLR meeting (short presentations):
Understanding deep learning requires rethinking generalization
Learning Features of Music from Scratch
Understanding intermediate layers using linear classifier probes
Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfer
Optimization as a model for few shot learning

31-3 Multimodal Compact Bilinear Pooling for Visual Question Answering and Visual Grounding Akira Fukui, Dong Huk Park, Daylen Yang, Anna Rohrbach, Trevor Darrell, Marcus Rohrbach

We also have a presentation of Marc Bolanos (UB).

24-3 Mask R-CNN Kaiming He, Georgia Gkioxari, Piotr Dollár, Ross Girshick

16-3 Perceptual Losses for Real-Time Style Transfer and Super-Resolution
Justin Johnson, Alexandre Alahi, Li Fei-Fei

9-3 PixelNet: Representation of the pixels, by the pixels, and for the pixels. Aayush Bansal, Xinlei Chen, Bryan Russell, Abhinav Gupta, Deva Ramanan

2-3 Presentation Juntung Pan (UPC) SalGAN: Visual Saliency Prediction with Generative Adversarial Networks

2-3 Unsupervised Pixel–Level Domain Adaptation with Generative Adversarial Networks Bousmalis, K., Silberman, N., Dohan, D., Erhan, D. and Krishnan, D.,

22-2 Recurrent Models of Visual Attention Volodymyr Mnih, Nicolas Heess, Alex Graves, Koray Kavukcuoglu

15-2 Instance-aware Semantic Segmentation via Multi-task Network Cascades Jifeng Dai Kaiming He Jian Sun

8-2 Highway Networks Rupesh Kumar Srivastava, Klaus Greff, Jurgen Schmidhuber

FractalNet: Ultra-Deep Neural Networks without Residuals Gustav Larsson, Michael Maire, Gregory Shakhnarovich

Densely Connected Convolutional Networks Gao Huang, Zhuang Liu, Kilian Q. Weinberger, Laurens van der Maaten

1-2 Unsupervised Domain Adaptation by Backpropagation, Yaroslav Ganin, Victor Lempitsky

25-1 Generating Videos with Scene Dynamics Carl Vondrick, Hamed Pirsiavash, Antonio Torralba

18-1 Generative Image Modeling using Style and Structure Adversarial Networks Xiaolong Wang, Abhinav Gupta


11-12** NIPS 2016 short presentations:**
Residual Networks Behave Like Ensembles of Relatively Shallow Networks
Training and Evaluating Multimodal Word Embeddings with Large-scale Web Annotated Images
Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling
Learning feed-forward one-shot learners
Visual Dynamics: Probabilistic Future Frame Synthesis via Cross Convolutional Networks
Domain Separation Networks
Coupled Generative Adversarial Networks
Learning the Number of Neurons in Deep Networks
Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networks
SoundNet: Learning Sound Representations from Unlabeled Video

29-11 NIPS 2016 short presentations:
Learning What and Where to Draw
R-FCN: Object Detection via Region-based Fully Convolutional Networks
Single-Image Depth Perception in the Wild
PerforatedCNNs: Acceleration through Elimination of Redundant Convolutions
CliqueCNN: Deep Unsupervised Exemplar Learning
Learning to learn by gradient descent by gradient descent
Conditional Image Generation with PixelCNN Decoders
Improved Techniques for Training GANs

23-11 Learning without Forgetting Zhizhong Li, Derek Hoiem

10-11 You only look once: Unified, real-time object detection Redmon, J., Divvala, S., Girshick, R. and Farhadi, A.,

SSD: Single shot multibox detector Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.Y. and Berg, A.C.

3-11 XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks Mohammad Rastegari, Vicente Ordonez, Joseph Redmon, Ali Farhadi

20-10 Learning local feature descriptors with triplets and shallow convolutional neural networks Vassileios Balntas, Edgar Riba, Daniel Ponsa and Krystian Mikolajczyk

6-10 Social LSTM: Human Trajectory Prediction in Crowded Spaces, A. Alahi, K. Goel, V. Ramanathan, A. Robicquet, L. Fei-Fei, S. Savarese

28-9 LSTM introduction (for example here).

21-9 Show and tell: A neural image caption generator Oriol Vinyals, Alexander Toshev, Samy Bengio, Dumitru Erhan

Sequence to Sequence Learning with Neural Networks Ilya Sutskever, Oriol Vinyals, Quoc V. Le

8-9 DenseCap: Fully Convolutional Localization Networks for Dense Captioning Justin Johnson, Andrej Karpathy, Li Fei-Fei

27-7 Fully-Convolutional Siamese Networks for Object Tracking Luca Bertinetto, Jack Valmadre, Joao F. Henriques, Andrea Vedaldi, Philip H. S. Torr

20-7 Generative adversarial text to image synthesis. Reed, Scott, Zeynep Akata, Xinchen Yan, Lajanugen Logeswaran, Bernt Schiele, and Honglak Lee.

13-7 Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun