A list of Computer Vision research papers that I found interesting.
I will continue adding more papers, whenever I come across something interesting.
- Paper on the "classical" approaches to image-based image synthesis is Photo Clip Art: http://graphics.cs.cmu.edu/projects/photoclipart/
-
ImageNet: A Large-Scale Hierarchical Image Database:
Website: http://www.image-net.org/
-
ImageNet Classification with Deep Convolutional Neural Networks. Alex Krizhevsky, Ilya Sutskever, Geoffrey E Hinton. NIPS 2012.
-
Interesting paper on learning how to design deep neural network architectures.
-
Sketch2Photo: Internet Image Montage.
Project Page: http://cg.cs.tsinghua.edu.cn/montage/main.htm
Paper: http://cg.cs.tsinghua.edu.cn/papers/SiggraphAsia_2009_sketch2photo.pdf
-
How Do Humans Sketch Objects?
Project page: http://cybertron.cg.tu-berlin.de/eitz/projects/classifysketch/
-
BIING: Binarized Normed Gradients for Objectness Estimation
Project page: http://mmcheng.net/bing/
-
What makes Paris look like Paris?
Project page: http://graphics.cs.cmu.edu/projects/whatMakesParis/
-
DeepFace: Closing the Gap to Human-Level Performance in Face Verification.
Paper: https://www.cs.toronto.edu/~ranzato/publications/taigman_cvpr14.pdf
-
Robust Video Segment Proposals with Painless Occlusion Handling.
Project Page: https://web.engr.oregonstate.edu/~lif/SegTrack2/Occlusion/
Paper: https://web.engr.oregonstate.edu/~lif/SegTrack2/Occlusion/occlusion_paper.pdf
-
DynamicFusion: Reconstruction and Tracking of Non-rigid Scenes in Real-Time
Project Page: http://grail.cs.washington.edu/projects/dynamicfusion/
Paper: http://grail.cs.washington.edu/projects/dynamicfusion/papers/DynamicFusion.pdf
-
Deep neural networks are easily fooled: High confidence predictions for unrecognizable images.
Project page: http://www.evolvingai.org/fooling
-
A Neural Algorithm of Artistic Style
-
Understanding deep features with computer-generated imagery
-
Learning Visual Biases from Human Imagination
-
FaceNet: A Unified Embedding for Face Recognition and Clustering
-
Joint Embeddings of Shapes and Images via CNN Image Purification
Paper: https://shapenet.cs.stanford.edu/projects/JointEmbedding/
-
The Sketchy Database: Learning to Retrieve Badly Drawn Bunnies
Project Page: http://sketchy.eye.gatech.edu/
-
Do Deep Convolutional Nets Really Need to be Deep and Convolutional?
-
Unsupervised Learning of Visual Representations using Videos
Paper: http://www.cs.cmu.edu/~xiaolonw/papers/unsupervised_video.pdf
Project Page: http://www.cs.cmu.edu/~xiaolonw/unsupervise.html
-
XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks
-
Deep Residual Learning for Image Recognition.
-
Learning Aligned Cross-Modal Representations from Weakly Aligned Data
Project Page: https://projects.csail.mit.edu/cmplaces/
-
Visually Indicated Sounds
vis.csail.mit.edu
-
"What happens if..." Learning to Predict the Effect of Forces in Images
-
YOLO9000: Better, Faster, Stronger
-
Visual Dialog
Paper: https://arxiv.org/abs/1611.08669
Project Page: http://visualdialog.org/
-
Generative Adversarial Nets
-
Unsupervised Representation Learning with Deep Convolutional GANs
-
Object Contour Detection with a Fully Convolutional Encoder-Decoder Network.
Project Page: https://eng.ucmerced.edu/people/jyang44/objectContourDetection.html
-
Photo Realistic Style Transfer - Recent work in transferring style between images.
-
Context Encoders: Feature Learning by Inpainting
-
Generative Adversarial Text-to-Image Synthesis
Project page - https://github.com/reedscot/icml2016
Paper - http://arxiv.org/abs/1605.05396
-
Generative Visual Manipulation on the Natural Image Manifold
Project page - http://people.eecs.berkeley.edu/~junyanz/projects/gvm/
-
Improved Techniques for Training GANs
Paper - https://arxiv.org/abs/1606.03498
-
Conditional Image Generation with PixelCNN Decoders
Paper - https://papers.nips.cc/paper/6527-conditional-image-generation-with-pixelcnn-decoders.pdf
Code - https://github.com/anantzoid/Conditional-PixelCNN-decoder
-
Attribute2Image: Conditional Image Generation from Visual Attributes
Paper: https://drive.google.com/file/d/0B9Q4vh7pPrOOdXpVTGZ6a1FlZW8/view
Project Page: https://sites.google.com/site/attribute2image/
Presentation Slides: https://docs.google.com/presentation/d/1qOaWY3qMviGRJm2QFMyYN3zgVcXQIEmmVNW56DiT_4I/edit#slide=id.g1d6af2990f_0_95
-
Semantic Segmentation using Adversarial Networks
Paper - https://arxiv.org/abs/1611.08408
-
NetVLAD: CNN architecture for weakly supervised place recognition.
paper: https://arxiv.org/abs/1511.07247
project page: http://www.di.ens.fr/willow/research/netvlad/
-
Image-to-Image Translation with Conditional Adversarial Nets.
Project - https://phillipi.github.io/pix2pix/
Have fun with the demo! - https://affinelayer.com/pixsrv/
Follow up work: CycleGAN - https://github.com/junyanz/CycleGAN
-
StackGAN: Text to Photo-realistic Image Synthesis with Stacked GANs
-
Neural Architecture Search with Reinforcement Learning
Paper: https://arxiv.org/abs/1611.01578
OpenReview Discussion: https://openreview.net/forum?id=r1Ue8Hcxg