Difference between revisions of "ReadingGroup"
m |
m |
||
(50 intermediate revisions by the same user not shown) | |||
Line 6: | Line 6: | ||
= Proposed Schedule = | = Proposed Schedule = | ||
+ | |||
+ | **Warning: This is now out of date. We are arranging via the email list now.** | ||
+ | |||
+ | <div style="opacity:0.2"> | ||
The below schedule is only '''proposed''', and subject to change. | The below schedule is only '''proposed''', and subject to change. | ||
Line 46: | Line 50: | ||
* <span style="color:#F09090">Week of 22 Janr: Break</span> | * <span style="color:#F09090">Week of 22 Janr: Break</span> | ||
* <span style="color:#F09090">Week of 29 Janr: Break</span> | * <span style="color:#F09090">Week of 29 Janr: Break</span> | ||
− | * <span style="color:# | + | * <span style="color:#000000"><s>Monday 05 Febr 3pm: Cao, Wu & Shen: Fully convolutional depth 1</s></span> |
− | * <span style="color:# | + | ** Location: EEBF 4302 |
− | * <span style="color:# | + | * <span style="color:#000000"><s>Monday 12 Febr 3:30pm: Laina et al.: Fully convolutional depth 2</s></span> |
− | * <span style="color:# | + | ** Location: EEBF 4302 |
− | * <span style="color:# | + | * <span style="color:#000000"><s>Monday 19 Febr 3:30pm: Li, Klein & Yao: Fully convolutional depth 3</s></span> |
− | * <span style="color:# | + | ** Location: EEBF 4302 |
− | * <span style="color:# | + | * <span style="color:#000000"><s>Monday 26 Febr 3:30pm: Güler et al.: DenseReg</s></span> |
− | * <span style="color:#F09090"> | + | ** Location: EEBF 4302 |
− | * <span style="color:# | + | * <span style="color:#000000"><s>Monday 05 Marc 3:30pm: Godard et al.: Unsupervised/train from stereo</s></span> |
− | * <span style="color:# | + | ** Location: EEBF 4302 |
− | * <span style="color:# | + | * <span style="color:#000000"><s>Monday 12 Marc 3:30pm: Zhou et al.: SfMLearner</s></span> |
− | * <span style="color:# | + | ** Location: EEBF 4302 |
− | * <span style="color:# | + | * <span style="color:#000000"><s>Monday 19 Marc 3:30pm: Fangchang & Karaman: depth from a single image & SLAM</s></span> |
− | * <span style="color:# | + | ** Location: EEBF 4302 |
− | * <span style="color:# | + | * <span style="color:#000000"><s>Monday 26 Marc 3:30pm: Pizzoli et al.: REMODE</s></span> |
− | * <span style="color:# | + | ** Location: EEBF 4302 |
− | * <span style="color:# | + | * <span style="color:#000000"><s>Monday 02 Aprl 3:30pm: Yan et al.: Superpixel CNN/CRF</s></span> |
− | * <span style="color:#CAAAAA">???: | + | ** Location: EEBF 4302 |
− | * <span style="color:#CAAAAA">???: Hoeim et al. Photo pop-up </span> | + | * <span style="color:#F09090">Monday 09 Aprl 3:30pm: Break</span> |
− | * <span style="color:#CAAAAA">???: Heitz et al. Cascaded Classification Models</span> | + | ** Location: EEBF 4302 |
− | * <span style="color:#CAAAAA">???: Li et al. Feedback-enabled Cascaded Classification Models</span> | + | * <span style="color:#000000">Monday 16 Aprl 3:30pm: Mnih et al.: Deep reinforcement learning</span> |
− | * <span style="color:#CAAAAA">???: Han et al. Bayesian object-level reconstruction</span> | + | ** Location: EEBF 4302 |
− | * <span style="color:#CAAAAA">???: Liu et al. Depth from semantics</span> | + | * <span style="color:#000000">Monday 23 Aprl 3:30pm: Dosovitskiy, Springenberg, Tatarchenko & Brox: Generating images</span> |
− | * <span style="color:#CAAAAA">???: Wu et al. Repetitive scene structure</span> | + | ** Location: EEBF 4302 |
− | * <span style="color:#CAAAAA">???: He et al. Haze removal</span> | + | * <span style="color:#000000">Monday 30 Aprl 3:30pm: Goodfellow et al.: Generative Adversarial Nets</span> |
− | * <span style="color:#CAAAAA">???: Hassner et al. Example-based Depth</span> | + | ** Location: EEBF 4302 |
+ | * <span style="color:#000000">Monday 07 May 3:30pm: Oord et al.: Pixel-RNN and Pixel-CNN</span> | ||
+ | ** Location: EEBF 4302 | ||
+ | * <span style="color:#000000">Monday 14 May 3:30pm: Isola et al.: Pix2Pix</span> | ||
+ | ** Location: EEBF 4302 | ||
+ | * <span style="color:#000000">Monday 21 May 3:30pm: Girshick, Donahue, Darrell & Malik: R-CNN</span> | ||
+ | ** Location: EEBF 4302 | ||
+ | * <span style="color:#CAAAAA">???: Redmon et al.: YOLO and YOLO9000</span> | ||
+ | * <span style="color:#CAAAAA">???: Khoreva et al.: Dense tracking/data augmentation</span> | ||
+ | * <span style="color:#CAAAAA">???m: Mancini et al.: Obstacle detection</span> | ||
+ | * <span style="color:#CAAAAA">???: Ilg/Fischer et al.: FlowNet/FlowNet 2.0</span> | ||
+ | * <span style="color:#CAAAAA">???: Pagnutti et al.: RGBD semantic segmentation with CNN + surface fitting</span> | ||
+ | * <span style="color:#CAAAAA">???: Zhu et al.: CycleGANs</span> | ||
+ | * <span style="color:#CAAAAA">???: Yang et al.: Full 3D reconstruction from single depth view</span> | ||
+ | * <span style="color:#CAAAAA">???: Li et al.: Fully Convolutional Instance-aware Semantic Segmentation</span> | ||
+ | * <span style="color:#CAAAAA">???: Kim et al.: Solving CRF with CNN (depth image)</span> | ||
+ | * <span style="color:#CAAAAA">???: Liu et al.: Attribute Grammar Scene Reconstruction</span> | ||
+ | * <span style="color:#CAAAAA">???: Tatarchenko et al.: Multi-view 3D models </span> | ||
+ | * <span style="color:#CAAAAA">???: Häne et al.: Single-view voxel reconstruction</span> | ||
+ | * <span style="color:#CAAAAA">???: Garg et al.: Geometry+CNN unsupervised</span> | ||
+ | * <span style="color:#CAAAAA">???: Xie et al.: Deep3D</span> | ||
+ | * <span style="color:#CAAAAA">???: Liu et al.: Convolutional Neural Field CRFs</span> | ||
+ | * <span style="color:#CAAAAA">???: Hong et al.: Semantic segmentation for robot behaviour</span> | ||
+ | * <span style="color:#CAAAAA">???: Liao, Huang, Wang, Kodagoda, Yu & Liu: Fuse with laser</span> | ||
+ | * <span style="color:#CAAAAA">???: Mirowski et al.: Learning to navigate</span> | ||
+ | * <span style="color:#CAAAAA">???: Finn & Levine: Visual prediction for planning</span> | ||
+ | * <span style="color:#CAAAAA">???: Roy & Todorovic: Neural Regression Forest</span> | ||
+ | * <span style="color:#CAAAAA">???: Hoeim et al.: Photo pop-up </span> | ||
+ | * <span style="color:#CAAAAA">???: Heitz et al.: Cascaded Classification Models</span> | ||
+ | * <span style="color:#CAAAAA">???: Li et al.: Feedback-enabled Cascaded Classification Models</span> | ||
+ | * <span style="color:#CAAAAA">???: Li et al.: Depth & Normals - CRF/regression</span> | ||
+ | * <span style="color:#CAAAAA">???: Han et al.: Bayesian object-level reconstruction</span> | ||
+ | * <span style="color:#CAAAAA">???: Liu et al.: Depth from semantics</span> | ||
+ | * <span style="color:#CAAAAA">???: Wu et al.: Repetitive scene structure</span> | ||
+ | * <span style="color:#CAAAAA">???: He et al.: Haze removal</span> | ||
+ | * <span style="color:#CAAAAA">???: Hassner et al.: Example-based Depth</span> | ||
+ | * <span style="color:#CAAAAA">???: Wu et al.: Repetition-based Depth</span> | ||
+ | |||
+ | </div> | ||
= Details = | = Details = | ||
+ | |||
+ | <div style="opacity:0.8"> | ||
== Foley & Maitlin Chapter 6 - Distance & Size Perception == | == Foley & Maitlin Chapter 6 - Distance & Size Perception == | ||
Line 242: | Line 286: | ||
Here we start a series of recent papers that take different approaches using deep nets to depth from a single image. | Here we start a series of recent papers that take different approaches using deep nets to depth from a single image. | ||
+ | |||
+ | Let me know if you can find some media for this work (I could only find the paper itself). | ||
== Laina et al.: Fully convolutional depth 2 == | == Laina et al.: Fully convolutional depth 2 == | ||
Line 266: | Line 312: | ||
http://visual.cs.ucl.ac.uk/pubs/monoDepth/ | http://visual.cs.ucl.ac.uk/pubs/monoDepth/ | ||
+ | |||
+ | https://github.com/mrharicot/monodepth | ||
== Zhou et al.: SfMLearner == | == Zhou et al.: SfMLearner == | ||
Line 274: | Line 322: | ||
https://arxiv.org/abs/1704.07813 | https://arxiv.org/abs/1704.07813 | ||
+ | |||
+ | A blog entry explaining the main ideas: http://bair.berkeley.edu/blog/2017/07/11/confluence-of-geometry-and-learning/ | ||
+ | |||
+ | The code: https://github.com/tinghuiz/SfMLearner | ||
+ | |||
+ | == Fangchang & Karaman: depth from a single image & SLAM == | ||
+ | |||
+ | http://www.mit.edu/~fcma/ | ||
+ | |||
+ | https://youtu.be/vNIIT_M7x7Y | ||
+ | |||
+ | https://arxiv.org/pdf/1709.07492.pdf | ||
+ | |||
+ | https://github.com/fangchangma/sparse-to-dense.git | ||
+ | |||
+ | == Pizzoli et al.: REMODE == | ||
+ | |||
+ | In case you miss it: this is not a single-image method... but close to it. It is another structure from motion method. But the results are rather good (state of the art in 2014). | ||
+ | |||
+ | [1]M. Pizzoli, C. Forster, and D. Scaramuzza, “REMODE: Probabilistic, monocular dense reconstruction in real time,” in Robotics and Automation (ICRA), 2014 IEEE International Conference on, 2014, pp. 2609–2616. | ||
+ | |||
+ | http://rpg.ifi.uzh.ch/docs/ICRA14_Pizzoli.pdf | ||
+ | |||
+ | https://www.youtube.com/watch?v=QTKd5UWCG0Q | ||
== Yan et al.: Superpixel CNN/CRF == | == Yan et al.: Superpixel CNN/CRF == | ||
Line 279: | Line 351: | ||
http://ieeexplore.ieee.org/document/8105853/ | http://ieeexplore.ieee.org/document/8105853/ | ||
− | == | + | == Mnih et al.: Deep reinforcement learning == |
+ | |||
+ | A modern classic. | ||
+ | |||
+ | http://arxiv.org/abs/1312.5602 | ||
+ | |||
+ | https://www.nature.com/articles/nature14236 (temporary link: http://web.itu.edu.tr/djduff/2018/nature14236.pdf ) | ||
− | https:// | + | https://www.youtube.com/watch?v=iqXKQf2BOSE |
− | Here | + | Here it is done in Keras: https://keon.io/deep-q-learning/ |
== Dosovitskiy, Springenberg, Tatarchenko & Brox: Generating images == | == Dosovitskiy, Springenberg, Tatarchenko & Brox: Generating images == | ||
https://arxiv.org/abs/1411.5928 | https://arxiv.org/abs/1411.5928 | ||
+ | |||
+ | https://ieeexplore.ieee.org/document/7469347/media | ||
A non-adversarial approach to generating images. | A non-adversarial approach to generating images. | ||
+ | |||
+ | Also see: | ||
+ | |||
+ | https://www.youtube.com/watch?v=QCSW4isBDL0 | ||
+ | |||
+ | https://www.youtube.com/watch?v=LAfmJQK4UW0 | ||
== Goodfellow et al.: Generative Adversarial Nets == | == Goodfellow et al.: Generative Adversarial Nets == | ||
Line 304: | Line 390: | ||
http://arxiv.org/abs/1606.05328 | http://arxiv.org/abs/1606.05328 | ||
+ | |||
+ | Some background on LSTMs: | ||
+ | |||
+ | http://colah.github.io/posts/2015-08-Understanding-LSTMs/ | ||
== Isola et al. Pix2Pix == | == Isola et al. Pix2Pix == | ||
Line 311: | Line 401: | ||
We can use this too. And it's cool. | We can use this too. And it's cool. | ||
− | + | The project page: | |
+ | |||
+ | https://phillipi.github.io/pix2pix/ | ||
− | + | An online demo: | |
− | + | https://affinelayer.com/pixsrv/ | |
== Girshick, Donahue, Darrell & Malik: R-CNN == | == Girshick, Donahue, Darrell & Malik: R-CNN == | ||
Line 321: | Line 413: | ||
https://arxiv.org/abs/1311.2524 | https://arxiv.org/abs/1311.2524 | ||
− | We take a slight segue to check out how tracking has been done recently with neural networks. Note that Faster-RCNN and more recent alternatives use similar principles but do it faster. | + | We take a slight segue to check out how tracking has been done recently with neural networks. Note that Faster-RCNN and more recent alternatives use similar principles but do it faster. Here we look at the older paper so that we can discuss some of the fundamentals. |
− | == Redmon et al.: YOLO == | + | == Redmon et al.: YOLO and YOLO9000 == |
+ | |||
+ | https://arxiv.org/abs/1612.08242 | ||
https://arxiv.org/abs/1506.02640 | https://arxiv.org/abs/1506.02640 | ||
− | == | + | == Khoreva et al.: Dense tracking/data augmentation == |
+ | |||
+ | An attempt on the DAVIS dataset. The dataset is here: http://davischallenge.org/ | ||
+ | |||
+ | This is interesting because of the data augmentation approach used. | ||
+ | |||
+ | The paper: http://arxiv.org/abs/1703.09554 | ||
− | https://arxiv.org/abs/1612. | + | Video: https://www.youtube.com/watch?v=QrsR5w-HR14 |
+ | |||
+ | == Mancini et al.: Obstacle detection == | ||
+ | |||
+ | https://arxiv.org/abs/1607.06349 | ||
+ | |||
+ | == Pagnutti et al. RGBD semantic segmentation with CNN + surface fitting == | ||
+ | |||
+ | http://ieeexplore.ieee.org/document/8120042/ | ||
+ | |||
+ | https://pdfs.semanticscholar.org/7716/9ee225157e77d1632e3bed54c70235b4abf0.pdf | ||
+ | |||
+ | == Zhu et al.: CycleGANs == | ||
+ | |||
+ | https://arxiv.org/abs/1703.10593 | ||
+ | |||
+ | Here is a nice tutorial: | ||
+ | |||
+ | https://hardikbansal.github.io/CycleGANBlog/ | ||
+ | |||
+ | == Ilg/Fischer et al.: FlowNet/FlowNet 2.0 == | ||
+ | |||
+ | FlowNet: | ||
+ | |||
+ | https://www.youtube.com/watch?v=g-peWXaQnQc | ||
+ | |||
+ | https://arxiv.org/abs/1504.06852 | ||
+ | |||
+ | FlowNet 2.0: | ||
+ | |||
+ | https://www.youtube.com/watch?v=JSzUdVBmQP4 | ||
+ | |||
+ | https://arxiv.org/abs/1612.01925 | ||
+ | |||
+ | == Yang et al.: Full 3D reconstruction from single depth view == | ||
+ | |||
+ | https://arxiv.org/abs/1708.07969 | ||
== Li et al. Fully Convolutional Instance-aware Semantic Segmentation == | == Li et al. Fully Convolutional Instance-aware Semantic Segmentation == | ||
https://arxiv.org/abs/1611.07709 | https://arxiv.org/abs/1611.07709 | ||
+ | |||
+ | == Kim et al. Solving CRF with CNN (depth image) == | ||
+ | |||
+ | http://arxiv.org/abs/1603.06359 | ||
+ | |||
+ | |||
+ | == Liu et al. Attribute Grammar Scene Reconstruction == | ||
+ | |||
+ | http://ieeexplore.ieee.org/document/7889053/?source=tocalert&dld=Z21haWwuY29t | ||
+ | |||
+ | == Tatarchenko et al. Multi-view 3D models == | ||
+ | |||
+ | https://arxiv.org/abs/1511.06702 | ||
+ | |||
+ | Not just inferring the depth image but also other views of it (related to the SfMLearner paper). | ||
+ | |||
+ | == Häne et al. Single-view voxel reconstruction == | ||
+ | |||
+ | Blog summary: http://bair.berkeley.edu/blog/2017/08/23/high-quality-3d-obj-reconstruction/ | ||
+ | |||
+ | Video intro: https://www.youtube.com/watch?v=BjwhMDhbqAs | ||
+ | |||
+ | Full paper: https://arxiv.org/abs/1704.00710 | ||
+ | |||
+ | == Garg et al. Geometry+CNN unsupervised == | ||
+ | |||
+ | We will have already read Godard et al. and Zhou et al. but this is for completeness. | ||
+ | |||
+ | http://arxiv.org/abs/1603.04992 | ||
+ | |||
+ | == Xie et al. Deep3D == | ||
+ | |||
+ | We will have already read Godard et al. and Zhou et al. but this is for completeness. | ||
+ | |||
+ | https://arxiv.org/pdf/1604.03650 | ||
+ | |||
+ | == Liu et al. Convolutional Neural Field CRFs == | ||
+ | |||
+ | https://arxiv.org/abs/1411.6387 | ||
+ | |||
+ | == Hong et al.: Semantic segmentation for robot behaviour == | ||
+ | |||
+ | https://arxiv.org/abs/1802.00285 | ||
+ | |||
+ | == Liao, Huang, Wang, Kodagoda, Yu & Liu: Fuse with laser == | ||
+ | |||
+ | https://arxiv.org/abs/1611.02174 | ||
+ | |||
+ | Here we see an interesting depth-from-single-image sensor fusion with robotics applications. | ||
+ | |||
+ | == Mirowski et al.: Learning to navigate == | ||
+ | |||
+ | https://arxiv.org/abs/1611.03673 | ||
+ | |||
+ | == Finn & Levine: Visual prediction for planning == | ||
+ | |||
+ | http://arxiv.org/abs/1610.00696 | ||
== Hoeim et al. Photo pop-up == | == Hoeim et al. Photo pop-up == | ||
Line 340: | Line 533: | ||
http://repository.cmu.edu/cgi/viewcontent.cgi?article=1288&context=robotics | http://repository.cmu.edu/cgi/viewcontent.cgi?article=1288&context=robotics | ||
+ | |||
+ | == Roy & Todorovic: Neural Regression Forest == | ||
+ | |||
+ | https://www.cv-foundation.org/openaccess/content_cvpr_2016/app/S23-11.pdf | ||
== Heitz et al. Cascaded Classification Models == | == Heitz et al. Cascaded Classification Models == | ||
Line 352: | Line 549: | ||
https://arxiv.org/abs/1110.5102 | https://arxiv.org/abs/1110.5102 | ||
+ | |||
+ | == Li et al. Depth & Normals - CRF/regression == | ||
+ | |||
+ | https://www.cv-foundation.org/openaccess/content_cvpr_2015/app/1B_001.pdf | ||
== Han et al. Bayesian object-level reconstruction == | == Han et al. Bayesian object-level reconstruction == | ||
Line 379: | Line 580: | ||
Seems like an older version of the SIFTFlow based one of Karsch. | Seems like an older version of the SIFTFlow based one of Karsch. | ||
− | + | https://www.researchgate.net/publication/4245893_Example_Based_3D_Reconstruction_from_Single_2D_Images | |
− | = | + | == Wu et al. Repetition-based Depth == |
− | |||
− | + | "Repetition-based dense single-view reconstruction" | |
− | + | ||
− | + | http://www.academia.edu/download/30713855/WuCVPR11.pdf | |
− | + | ||
− | + | </div> | |
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
= Additional Resources = | = Additional Resources = | ||
Line 421: | Line 597: | ||
My NN/Keras bootcamp slides: http://files.djduff.net/nn.zip | My NN/Keras bootcamp slides: http://files.djduff.net/nn.zip | ||
+ | |||
+ | Foley & Maitlin's book: https://books.google.com.tr/books?id=jLBmCgAAQBAJ&printsec=frontcover |
Latest revision as of 05:38, 6 September 2018
Contents
- 1 Getting involved
- 2 Proposed Schedule
- 3 Details
- 3.1 Foley & Maitlin Chapter 6 - Distance & Size Perception
- 3.2 Saxena, Min & Ng: Make3D
- 3.3 Michels, Saxena & Ng: High speed obstacle avoidance
- 3.4 Karsch, Liu & Kang: Depth Transfer
- 3.5 LeCun, Bengio & Hinton: Deep Learning Review
- 3.6 Rumelhart, Hinton & Williams: Backpropagation
- 3.7 LeCun, Bottou, Bengio & Haffner: CNNs
- 3.8 Krizhevsky, Sutskever & Hinton: ImageNet/AlexNet
- 3.9 Simonyan & Zisserman: VGG-16
- 3.10 Eigen, Puhrsch & Fergus: Depth map prediction
- 3.11 Luo et al.: Deep Learning + Stereo
- 3.12 Shelhamer, Long & Darrell: Fully Convolutional Segmentation
- 3.13 Giuisti et al.: Forest trails CNN
- 3.14 Loffe & Szegedy: Batch Normalization
- 3.15 He, Zhang, Ren & Sun: ResNets
- 3.16 Cao, Wu & Shen: Fully convolutional depth 1
- 3.17 Laina et al.: Fully convolutional depth 2
- 3.18 Li, Klein & Yao: Fully convolutional depth 3
- 3.19 Güler et al. DenseReg
- 3.20 Godard et al. Unsupervised/train from stereo
- 3.21 Zhou et al.: SfMLearner
- 3.22 Fangchang & Karaman: depth from a single image & SLAM
- 3.23 Pizzoli et al.: REMODE
- 3.24 Yan et al.: Superpixel CNN/CRF
- 3.25 Mnih et al.: Deep reinforcement learning
- 3.26 Dosovitskiy, Springenberg, Tatarchenko & Brox: Generating images
- 3.27 Goodfellow et al.: Generative Adversarial Nets
- 3.28 Oord et al.: Pixel-RNN & Pixel-CNN
- 3.29 Isola et al. Pix2Pix
- 3.30 Girshick, Donahue, Darrell & Malik: R-CNN
- 3.31 Redmon et al.: YOLO and YOLO9000
- 3.32 Khoreva et al.: Dense tracking/data augmentation
- 3.33 Mancini et al.: Obstacle detection
- 3.34 Pagnutti et al. RGBD semantic segmentation with CNN + surface fitting
- 3.35 Zhu et al.: CycleGANs
- 3.36 Ilg/Fischer et al.: FlowNet/FlowNet 2.0
- 3.37 Yang et al.: Full 3D reconstruction from single depth view
- 3.38 Li et al. Fully Convolutional Instance-aware Semantic Segmentation
- 3.39 Kim et al. Solving CRF with CNN (depth image)
- 3.40 Liu et al. Attribute Grammar Scene Reconstruction
- 3.41 Tatarchenko et al. Multi-view 3D models
- 3.42 Häne et al. Single-view voxel reconstruction
- 3.43 Garg et al. Geometry+CNN unsupervised
- 3.44 Xie et al. Deep3D
- 3.45 Liu et al. Convolutional Neural Field CRFs
- 3.46 Hong et al.: Semantic segmentation for robot behaviour
- 3.47 Liao, Huang, Wang, Kodagoda, Yu & Liu: Fuse with laser
- 3.48 Mirowski et al.: Learning to navigate
- 3.49 Finn & Levine: Visual prediction for planning
- 3.50 Hoeim et al. Photo pop-up
- 3.51 Roy & Todorovic: Neural Regression Forest
- 3.52 Heitz et al. Cascaded Classification Models
- 3.53 Li et al. Feedback-enabled Cascaded Classification Models
- 3.54 Li et al. Depth & Normals - CRF/regression
- 3.55 Han et al. Bayesian object-level reconstruction
- 3.56 Liu et al. Depth from semantics
- 3.57 Wu et al. Repetitive scene structure
- 3.58 He et al. Haze removal
- 3.59 Hassner et al. Example-based Depth
- 3.60 Wu et al. Repetition-based Depth
- 4 Additional Resources
Getting involved
There is now a mailing list for this reading group. Send me (Damien) an email to get on it. No problem.
Note: this reading group is about deep learning as applied to depth estimation from a single image - one of the super hot topics. If your interest is deep learning in general, you may find some of the readings a little bit off-topic. So let me know if you want some idea about which you should read for.
Proposed Schedule
- Warning: This is now out of date. We are arranging via the email list now.**
The below schedule is only proposed, and subject to change.
- 14 Sept 8.30am:
Foley & Maitlin Chapter 6: Distance & Size Perception- Location: EEBF 4302
- 21 Sept 8.30am:
Saxena, Min & Ng: Make3D- Location: EEBF 4302
- 28 Sept 8.30am:
Michels, Saxena & Ng: High speed obstacle avoidance- Location: EEBF 4302
- 05 Octr 8.30am:
Karsch, Liu & Kang: Depth Transfer- Location: EEBF 4302
- 12 Octr 8.30am:
LeCun, Bengio & Hinton: Deep Learning Review- Location: EEBF 4302
- 19 Octr 8.30am:
Rumelhart, Hinton & Williams: Backpropagation- Location: EEBF 4302
- 26 Octr 8.30am:
LeCun, Bottou, Bengio & Haffner: CNNs- Location: EEBF 4302
- 02 Novr 8:30am:
LeCun, Bottou, Bengio & Haffner: CNNs- Location: EEBF 4302
- 09 Novr 8.30am:
Simonyan & Zisserman: VGG-16- Location: EEBF 4302
- 16 Novr 8.30am:
Eigen, Puhrsch & Fergus: Depth map prediction- Location: EEBF 4302
- 23 Novr 8.30am:
Luo et al.: Deep Learning + Stereo- Location: EEBF 4302
- 30 Novr 8.30am:
Shelhamer, Long & Darrell: Fully Convolutional Segmentation- Location: EEBF 4302
- 07 Decr 8.30am:
Giuisti et al.: Forest trails CNN- Location: EEBF 4302
- 14 Decr 8.30am:
Batch Normalization- Location: EEBF 4302
- 22 Decr 8:30am:
He, Zhang, Ren & Sun: ResNet- Location: EEBF 4302
- Week of 25 Decr: Break
- Week of 01 Janr: Break
- Week of 08 Janr: Break
- Week of 15 Janr: Break
- Week of 22 Janr: Break
- Week of 29 Janr: Break
-
Monday 05 Febr 3pm: Cao, Wu & Shen: Fully convolutional depth 1- Location: EEBF 4302
-
Monday 12 Febr 3:30pm: Laina et al.: Fully convolutional depth 2- Location: EEBF 4302
-
Monday 19 Febr 3:30pm: Li, Klein & Yao: Fully convolutional depth 3- Location: EEBF 4302
-
Monday 26 Febr 3:30pm: Güler et al.: DenseReg- Location: EEBF 4302
-
Monday 05 Marc 3:30pm: Godard et al.: Unsupervised/train from stereo- Location: EEBF 4302
-
Monday 12 Marc 3:30pm: Zhou et al.: SfMLearner- Location: EEBF 4302
-
Monday 19 Marc 3:30pm: Fangchang & Karaman: depth from a single image & SLAM- Location: EEBF 4302
-
Monday 26 Marc 3:30pm: Pizzoli et al.: REMODE- Location: EEBF 4302
-
Monday 02 Aprl 3:30pm: Yan et al.: Superpixel CNN/CRF- Location: EEBF 4302
- Monday 09 Aprl 3:30pm: Break
- Location: EEBF 4302
- Monday 16 Aprl 3:30pm: Mnih et al.: Deep reinforcement learning
- Location: EEBF 4302
- Monday 23 Aprl 3:30pm: Dosovitskiy, Springenberg, Tatarchenko & Brox: Generating images
- Location: EEBF 4302
- Monday 30 Aprl 3:30pm: Goodfellow et al.: Generative Adversarial Nets
- Location: EEBF 4302
- Monday 07 May 3:30pm: Oord et al.: Pixel-RNN and Pixel-CNN
- Location: EEBF 4302
- Monday 14 May 3:30pm: Isola et al.: Pix2Pix
- Location: EEBF 4302
- Monday 21 May 3:30pm: Girshick, Donahue, Darrell & Malik: R-CNN
- Location: EEBF 4302
- ???: Redmon et al.: YOLO and YOLO9000
- ???: Khoreva et al.: Dense tracking/data augmentation
- ???m: Mancini et al.: Obstacle detection
- ???: Ilg/Fischer et al.: FlowNet/FlowNet 2.0
- ???: Pagnutti et al.: RGBD semantic segmentation with CNN + surface fitting
- ???: Zhu et al.: CycleGANs
- ???: Yang et al.: Full 3D reconstruction from single depth view
- ???: Li et al.: Fully Convolutional Instance-aware Semantic Segmentation
- ???: Kim et al.: Solving CRF with CNN (depth image)
- ???: Liu et al.: Attribute Grammar Scene Reconstruction
- ???: Tatarchenko et al.: Multi-view 3D models
- ???: Häne et al.: Single-view voxel reconstruction
- ???: Garg et al.: Geometry+CNN unsupervised
- ???: Xie et al.: Deep3D
- ???: Liu et al.: Convolutional Neural Field CRFs
- ???: Hong et al.: Semantic segmentation for robot behaviour
- ???: Liao, Huang, Wang, Kodagoda, Yu & Liu: Fuse with laser
- ???: Mirowski et al.: Learning to navigate
- ???: Finn & Levine: Visual prediction for planning
- ???: Roy & Todorovic: Neural Regression Forest
- ???: Hoeim et al.: Photo pop-up
- ???: Heitz et al.: Cascaded Classification Models
- ???: Li et al.: Feedback-enabled Cascaded Classification Models
- ???: Li et al.: Depth & Normals - CRF/regression
- ???: Han et al.: Bayesian object-level reconstruction
- ???: Liu et al.: Depth from semantics
- ???: Wu et al.: Repetitive scene structure
- ???: He et al.: Haze removal
- ???: Hassner et al.: Example-based Depth
- ???: Wu et al.: Repetition-based Depth
Details
Foley & Maitlin Chapter 6 - Distance & Size Perception
Because our project is about using machine learning to extract depth from a single image (with deep learning, then applying it to robot problems) it pays to learn a bit about how humans do it...
https://books.google.com.tr/books?id=jLBmCgAAQBAJ&printsec=frontcover
Go to Chapter 6.
If that doesn't work (some have reported finding it difficult to access Chapter 6), try the following link: http://tinyurl.com/yalnnwp9 - some have reported being able to access the chapter by doing a google search for content.
Another thing to try that has worked for some is to log out of any google/gmail account before trying to access.
If nothing else works, email me.
Saxena, Min & Ng: Make3D
http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4531745
This is the classic paper that brought machine learning to the problem of depth from a single image, quite successfully, considering previous attempts. It uses Markov Random Fields, which are a bit advanced but, importantly, quite slow.
Note: because our library has a subscription to IEEE Xplore, you can access the above link from on-campus or via off-campus library access or via VPN.
But, here is an alternative link: http://www.cs.cornell.edu/~asaxena/reconstruction3d/saxena_make3d_learning3dstructure.pdf
There are some videos and things available here: http://make3d.cs.cornell.edu/ -- there used to be a live online demo but they've closed that. There is also a list of results on the Make3D dataset up till about 2012: http://make3d.cs.cornell.edu/results_stateoftheart.html
After that other datasets started being used also.
Superpixels are used in the study. Here is a quick intro to them: http://ttic.uchicago.edu/~xren/research/superpixel/
MRFs are more difficult and if anybody has seen a good tutorial for them let me know so that I can link to it here. The best I could find is https://mitpress.mit.edu/sites/default/files/titles/content/9780262015776_sch_0001.pdf but it is still a bit difficult. We will probably end up discussing what MRFs are a lot on Thursday.
Michels, Saxena & Ng: High speed obstacle avoidance
http://dl.acm.org/citation.cfm?id=1102426
Here the same authors focus on a related problem, that of determining open spaces for guiding a vehicle, again using machine learning techniques.
This version of the paper might be of higher quality (thanks to Hossein for finding):
http://ai.stanford.edu/~asaxena/rccar/ICML_ObstacleAvoidance.pdf
Karsch, Liu & Kang: Depth Transfer
Here is the target paper: http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=5551153
For those who are not on campus, a temporary link: http://web.itu.edu.tr/djduff/Share/KarschEtAl2014.pdf
This is a nonparametric approach to depth from a single image. They search a database of images similar to the observed one then aligns the found image with the observed one then warps the found image retrieved from the database to estimate the depth of the current image. It depends on an approach called SIFTFlow to do the alignment.
Here is a paper describing "SIFTFlow" (if you have the time to go deeper): http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=6787109
Free version: http://people.csail.mit.edu/celiu/SIFTflow/
Or a shorter conference version available off-campus: http://people.csail.mit.edu/celiu/ECCV2008/
LeCun, Bengio & Hinton: Deep Learning Review
http://www.nature.com/nature/journal/v521/n7553/full/nature14539.html?foxtrotcallback=true
Alternative links: http://pages.cs.wisc.edu/~dyer/cs540/handouts/deep-learning-nature2015.pdf https://www.researchgate.net/publication/277411157_Deep_Learning
A whirlwind compressed intro to deep learning and its parts.
For a more gentle introduction to deep learning: http://cs231n.stanford.edu/
Or you can find lots of gentle short intros: https://www.google.com.tr/search?q=intro+to+deep+learning
Rumelhart, Hinton & Williams: Backpropagation
An early paper introducing backpropagation, the main way we train neural networks nowadays: http://www.nature.com/articles/323533a0
Alternative link: http://www.cs.toronto.edu/~hinton/absps/naturebp.pdf
These topics are also addressed in the tutorials shared just above (or you will find plenty online and most neural network tutorials attempt to explain backpropagation as it is the main way these networks are trained - I usually use a simple genetic algorithm when explaining how to train neural networks because it's simpler - an accelerated tutorial of neural networks without explaining backpropagation but explaining one of the tools is at http://files.djduff.net/nn.zip ).
LeCun, Bottou, Bengio & Haffner: CNNs
This is the now classic paper describing LeNet architectures applying Convolutional Neural Networks (CNNs) to the problem of optical character recognition. It also embeds the neural network in an architecture for automatically segmenting text, including a system for automatically reading cheques.
http://ieeexplore.ieee.org/abstract/document/726791/
Alternative link: http://www.dengfanxin.cn/wp-content/uploads/2016/03/1998Lecun.pdf
Convolutional neural networks were introduced by the authors in 1990. It may be instructive to read that considerably simpler paper: http://yann.lecun.com/exdb/publis/pdf/lecun-90c.pdf
Or any tutorial about CNNs. One good place to follow is: http://cs231n.stanford.edu/syllabus.html
Krizhevsky, Sutskever & Hinton: ImageNet/AlexNet
We will not discuss this in the reading group.
Here is when convolutional neural networks and deep learning really showed what it could do - the problem of image recognition.
But we won't use this because a lot of the complexity it introduces turns out not to be necessary. Later methods are "cleaner". So we have taken it out of the reading list.
Simonyan & Zisserman: VGG-16
http://arxiv.org/abs/1409.1556
A relatively recent "deep" deep net with 16 layers for image recognition. Note: successful recent networks have one thousand layers.
Feel free to take a look at AlexNet above to get an idea of the space of approaches.
Eigen, Puhrsch & Fergus: Depth map prediction
https://www.cs.nyu.edu/~deigen/depth/
Finally, we apply deep neural convolutional networks to the problem that we are interested in.
Luo et al.: Deep Learning + Stereo
Combining deep learning and stereo.
https://www.cs.toronto.edu/~urtasun/publications/luo_etal_cvpr16.pdf
Shelhamer, Long & Darrell: Fully Convolutional Segmentation
http://arxiv.org/abs/1605.06211
Here a related problem is solved, that of semantic segmentation, but this approach is applicable to our problem.
Giuisti et al.: Forest trails CNN
http://ieeexplore.ieee.org/document/7358076/
Alternative link: http://rpg.ifi.uzh.ch/docs/RAL16_Giusti.pdf
See also youtube: https://www.youtube.com/watch?v=umRdt3zGgpU
Here we have a CNN-based update to the learn-to-navigate-from-images problem addressed by Saxena et al. above.
Loffe & Szegedy: Batch Normalization
https://arxiv.org/abs/1502.03167
A recent technique that has enabled powerful new methods and ultimately much deeper neural networks. Important stuff.
Apparently the following video from the Stanford CS231N course contains details of Batch Normalization (37:00 to 59:30)
https://youtu.be/gYpoJMlgyXA?list=PLkt2uSq6rBVctENoVBg1TpCC7OQi31AlC
Best, Damien
He, Zhang, Ren & Sun: ResNets
https://arxiv.org/abs/1512.03385
This work and variations on it have been the basis of the 1000 layer recent neural networks. Important stuff.
The diagram at the top of Page 2 of this paper (which offers an improvement on the ResNet of the above paper) is quite useful in understanding the structure of a residual unit:
http://arxiv.org/abs/1603.05027
Cao, Wu & Shen: Fully convolutional depth 1
http://arxiv.org/abs/1605.02305
Here we start a series of recent papers that take different approaches using deep nets to depth from a single image.
Let me know if you can find some media for this work (I could only find the paper itself).
Laina et al.: Fully convolutional depth 2
http://arxiv.org/abs/1606.00373
Here we continue a series of recent papers that take different approaches using deep nets to depth from a single image.
Li, Klein & Yao: Fully convolutional depth 3
http://arxiv.org/abs/1607.00730
Here we finalise a series of recent papers that take different approaches using deep nets to depth from a single image.
Güler et al. DenseReg
https://arxiv.org/abs/1612.01202
A key idea here is how to do regression using categorical prediction & an application of the Fully Convolutional Networks to regression problems.
Godard et al. Unsupervised/train from stereo
https://arxiv.org/abs/1609.03677
http://visual.cs.ucl.ac.uk/pubs/monoDepth/
https://github.com/mrharicot/monodepth
Zhou et al.: SfMLearner
Super cool stuff.
https://people.eecs.berkeley.edu/%7Etinghuiz/projects/SfMLearner/
https://arxiv.org/abs/1704.07813
A blog entry explaining the main ideas: http://bair.berkeley.edu/blog/2017/07/11/confluence-of-geometry-and-learning/
The code: https://github.com/tinghuiz/SfMLearner
Fangchang & Karaman: depth from a single image & SLAM
https://arxiv.org/pdf/1709.07492.pdf
https://github.com/fangchangma/sparse-to-dense.git
Pizzoli et al.: REMODE
In case you miss it: this is not a single-image method... but close to it. It is another structure from motion method. But the results are rather good (state of the art in 2014).
[1]M. Pizzoli, C. Forster, and D. Scaramuzza, “REMODE: Probabilistic, monocular dense reconstruction in real time,” in Robotics and Automation (ICRA), 2014 IEEE International Conference on, 2014, pp. 2609–2616.
http://rpg.ifi.uzh.ch/docs/ICRA14_Pizzoli.pdf
https://www.youtube.com/watch?v=QTKd5UWCG0Q
Yan et al.: Superpixel CNN/CRF
http://ieeexplore.ieee.org/document/8105853/
Mnih et al.: Deep reinforcement learning
A modern classic.
http://arxiv.org/abs/1312.5602
https://www.nature.com/articles/nature14236 (temporary link: http://web.itu.edu.tr/djduff/2018/nature14236.pdf )
https://www.youtube.com/watch?v=iqXKQf2BOSE
Here it is done in Keras: https://keon.io/deep-q-learning/
Dosovitskiy, Springenberg, Tatarchenko & Brox: Generating images
https://arxiv.org/abs/1411.5928
https://ieeexplore.ieee.org/document/7469347/media
A non-adversarial approach to generating images.
Also see:
https://www.youtube.com/watch?v=QCSW4isBDL0
https://www.youtube.com/watch?v=LAfmJQK4UW0
Goodfellow et al.: Generative Adversarial Nets
https://papers.nips.cc/paper/5423-generative-adversarial-nets
Another important recent development that we may make use of.
Oord et al.: Pixel-RNN & Pixel-CNN
https://arxiv.org/abs/1601.06759
Producing distributions over images. We have always intended to do something like this for depth images.
http://arxiv.org/abs/1606.05328
Some background on LSTMs:
http://colah.github.io/posts/2015-08-Understanding-LSTMs/
Isola et al. Pix2Pix
https://arxiv.org/abs/1611.07004
We can use this too. And it's cool.
The project page:
https://phillipi.github.io/pix2pix/
An online demo:
https://affinelayer.com/pixsrv/
Girshick, Donahue, Darrell & Malik: R-CNN
https://arxiv.org/abs/1311.2524
We take a slight segue to check out how tracking has been done recently with neural networks. Note that Faster-RCNN and more recent alternatives use similar principles but do it faster. Here we look at the older paper so that we can discuss some of the fundamentals.
Redmon et al.: YOLO and YOLO9000
https://arxiv.org/abs/1612.08242
https://arxiv.org/abs/1506.02640
Khoreva et al.: Dense tracking/data augmentation
An attempt on the DAVIS dataset. The dataset is here: http://davischallenge.org/
This is interesting because of the data augmentation approach used.
The paper: http://arxiv.org/abs/1703.09554
Video: https://www.youtube.com/watch?v=QrsR5w-HR14
Mancini et al.: Obstacle detection
https://arxiv.org/abs/1607.06349
Pagnutti et al. RGBD semantic segmentation with CNN + surface fitting
http://ieeexplore.ieee.org/document/8120042/
https://pdfs.semanticscholar.org/7716/9ee225157e77d1632e3bed54c70235b4abf0.pdf
Zhu et al.: CycleGANs
https://arxiv.org/abs/1703.10593
Here is a nice tutorial:
https://hardikbansal.github.io/CycleGANBlog/
Ilg/Fischer et al.: FlowNet/FlowNet 2.0
FlowNet:
https://www.youtube.com/watch?v=g-peWXaQnQc
https://arxiv.org/abs/1504.06852
FlowNet 2.0:
https://www.youtube.com/watch?v=JSzUdVBmQP4
https://arxiv.org/abs/1612.01925
Yang et al.: Full 3D reconstruction from single depth view
https://arxiv.org/abs/1708.07969
Li et al. Fully Convolutional Instance-aware Semantic Segmentation
https://arxiv.org/abs/1611.07709
Kim et al. Solving CRF with CNN (depth image)
http://arxiv.org/abs/1603.06359
Liu et al. Attribute Grammar Scene Reconstruction
http://ieeexplore.ieee.org/document/7889053/?source=tocalert&dld=Z21haWwuY29t
Tatarchenko et al. Multi-view 3D models
https://arxiv.org/abs/1511.06702
Not just inferring the depth image but also other views of it (related to the SfMLearner paper).
Häne et al. Single-view voxel reconstruction
Blog summary: http://bair.berkeley.edu/blog/2017/08/23/high-quality-3d-obj-reconstruction/
Video intro: https://www.youtube.com/watch?v=BjwhMDhbqAs
Full paper: https://arxiv.org/abs/1704.00710
Garg et al. Geometry+CNN unsupervised
We will have already read Godard et al. and Zhou et al. but this is for completeness.
http://arxiv.org/abs/1603.04992
Xie et al. Deep3D
We will have already read Godard et al. and Zhou et al. but this is for completeness.
https://arxiv.org/pdf/1604.03650
Liu et al. Convolutional Neural Field CRFs
https://arxiv.org/abs/1411.6387
Hong et al.: Semantic segmentation for robot behaviour
https://arxiv.org/abs/1802.00285
Liao, Huang, Wang, Kodagoda, Yu & Liu: Fuse with laser
https://arxiv.org/abs/1611.02174
Here we see an interesting depth-from-single-image sensor fusion with robotics applications.
https://arxiv.org/abs/1611.03673
Finn & Levine: Visual prediction for planning
http://arxiv.org/abs/1610.00696
Hoeim et al. Photo pop-up
A classic.
http://repository.cmu.edu/cgi/viewcontent.cgi?article=1288&context=robotics
Roy & Todorovic: Neural Regression Forest
https://www.cv-foundation.org/openaccess/content_cvpr_2016/app/S23-11.pdf
Heitz et al. Cascaded Classification Models
Older pre-CNN machine learning papers for depth estimation from a single image.
Li et al. Feedback-enabled Cascaded Classification Models
Older pre-CNN machine learning papers for depth estimation from a single image.
https://arxiv.org/abs/1110.5102
Li et al. Depth & Normals - CRF/regression
https://www.cv-foundation.org/openaccess/content_cvpr_2015/app/1B_001.pdf
Han et al. Bayesian object-level reconstruction
http://escholarship.org/uc/item/9tk6935x.pdf
Liu et al. Depth from semantics
Older pre-CNN machine learning papers for depth estimation from a single image.
http://ai.stanford.edu/people/koller/Papers/Liu+al:CVPR10.pdf
Wu et al. Repetitive scene structure
Older pre-CNN machine learning papers for depth estimation from a single image.
http://www.academia.edu/download/30713855/WuCVPR11.pdf
He et al. Haze removal
Might be interesting because of use of single-image cues.
http://mmlab.ie.cuhk.edu.hk/2009/dehaze_cvpr2009.pdf
Hassner et al. Example-based Depth
Seems like an older version of the SIFTFlow based one of Karsch.
Wu et al. Repetition-based Depth
"Repetition-based dense single-view reconstruction"
Additional Resources
EBook: https://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/DeepLearning-NowPublishing-Vol7-SIG-039.pdf Deep Learning: Methods and Applications by Li Deng and Dong Yu
Online course with slides videos and assignments: http://cs231n.stanford.edu/ CS231n: Convolutional Neural Networks for Visual Recognition (Stanford)
My NN/Keras bootcamp slides: http://files.djduff.net/nn.zip
Foley & Maitlin's book: https://books.google.com.tr/books?id=jLBmCgAAQBAJ&printsec=frontcover