MediaWiki API result
This is the HTML representation of the JSON format. HTML is good for debugging, but is unsuitable for application use.
Specify the format parameter to change the output format. To see the non-HTML representation of the JSON format, set format=json.
See the complete documentation, or the API help for more information.
{
"batchcomplete": "",
"continue": {
"gapcontinue": "Slot_Online",
"continue": "gapcontinue||"
},
"query": {
"pages": {
"3": {
"pageid": 3,
"ns": 0,
"title": "ReadingGroup",
"revisions": [
{
"contentformat": "text/x-wiki",
"contentmodel": "wikitext",
"*": "= Getting involved =\n\nThere is now a mailing list for this reading group. Send me (Damien) an email to get on it. No problem.\n\n'''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let me know if you want some idea about which you should read for.\n\n= Proposed Schedule =\n\n**Warning: This is now out of date. We are arranging via the email list now.**\n\n<div style=\"opacity:0.2\">\n\nThe below schedule is only '''proposed''', and subject to change.\n\n* 14 Sept 8.30am: <s>Foley & Maitlin Chapter 6: Distance & Size Perception</s>\n** Location: EEBF 4302\n* 21 Sept 8.30am: <s>Saxena, Min & Ng: Make3D</s>\n** Location: EEBF 4302\n* 28 Sept 8.30am: <s>Michels, Saxena & Ng: High speed obstacle avoidance</s>\n** Location: EEBF 4302\n* 05 Octr 8.30am: <s>Karsch, Liu & Kang: Depth Transfer</s>\n** Location: EEBF 4302\n* 12 Octr 8.30am: <s>LeCun, Bengio & Hinton: Deep Learning Review</s>\n** Location: EEBF 4302\n* 19 Octr 8.30am: <s>Rumelhart, Hinton & Williams: Backpropagation</s>\n** Location: EEBF 4302\n* 26 Octr 8.30am: <s>LeCun, Bottou, Bengio & Haffner: CNNs</s>\n** Location: EEBF 4302\n* 02 Novr 8:30am: <s>LeCun, Bottou, Bengio & Haffner: CNNs</s>\n** Location: EEBF 4302\n* 09 Novr 8.30am: <s>Simonyan & Zisserman: VGG-16</s>\n** Location: EEBF 4302\n* 16 Novr 8.30am: <s>Eigen, Puhrsch & Fergus: Depth map prediction</s>\n** Location: EEBF 4302\n* 23 Novr 8.30am: <s>Luo et al.: Deep Learning + Stereo</s>\n** Location: EEBF 4302\n* 30 Novr 8.30am: <s>Shelhamer, Long & Darrell: Fully Convolutional Segmentation</s>\n** Location: EEBF 4302\n* 07 Decr 8.30am: <s>Giuisti et al.: Forest trails CNN</s>\n** Location: EEBF 4302\n* 14 Decr 8.30am: <s>Batch Normalization</s>\n** Location: EEBF 4302\n* 22 Decr 8:30am: <s>He, Zhang, Ren & Sun: ResNet</s>\n** Location: EEBF 4302\n\n* <span style=\"color:#F09090\">Week of 25 Decr: Break</span>\n* <span style=\"color:#F09090\">Week of 01 Janr: Break</span>\n* <span style=\"color:#F09090\">Week of 08 Janr: Break</span>\n* <span style=\"color:#F09090\">Week of 15 Janr: Break</span>\n* <span style=\"color:#F09090\">Week of 22 Janr: Break</span>\n* <span style=\"color:#F09090\">Week of 29 Janr: Break</span>\n* <span style=\"color:#000000\"><s>Monday 05 Febr 3pm: Cao, Wu & Shen: Fully convolutional depth 1</s></span>\n** Location: EEBF 4302\n* <span style=\"color:#000000\"><s>Monday 12 Febr 3:30pm: Laina et al.: Fully convolutional depth 2</s></span>\n** Location: EEBF 4302\n* <span style=\"color:#000000\"><s>Monday 19 Febr 3:30pm: Li, Klein & Yao: Fully convolutional depth 3</s></span>\n** Location: EEBF 4302\n* <span style=\"color:#000000\"><s>Monday 26 Febr 3:30pm: G\u00fcler et al.: DenseReg</s></span>\n** Location: EEBF 4302\n* <span style=\"color:#000000\"><s>Monday 05 Marc 3:30pm: Godard et al.: Unsupervised/train from stereo</s></span>\n** Location: EEBF 4302\n* <span style=\"color:#000000\"><s>Monday 12 Marc 3:30pm: Zhou et al.: SfMLearner</s></span>\n** Location: EEBF 4302\n* <span style=\"color:#000000\"><s>Monday 19 Marc 3:30pm: Fangchang & Karaman: depth from a single image & SLAM</s></span>\n** Location: EEBF 4302\n* <span style=\"color:#000000\"><s>Monday 26 Marc 3:30pm: Pizzoli et al.: REMODE</s></span>\n** Location: EEBF 4302\n* <span style=\"color:#000000\"><s>Monday 02 Aprl 3:30pm: Yan et al.: Superpixel CNN/CRF</s></span>\n** Location: EEBF 4302\n* <span style=\"color:#F09090\">Monday 09 Aprl 3:30pm: Break</span>\n** Location: EEBF 4302\n* <span style=\"color:#000000\">Monday 16 Aprl 3:30pm: Mnih et al.: Deep reinforcement learning</span>\n** Location: EEBF 4302\n* <span style=\"color:#000000\">Monday 23 Aprl 3:30pm: Dosovitskiy, Springenberg, Tatarchenko & Brox: Generating images</span>\n** Location: EEBF 4302\n* <span style=\"color:#000000\">Monday 30 Aprl 3:30pm: Goodfellow et al.: Generative Adversarial Nets</span>\n** Location: EEBF 4302\n* <span style=\"color:#000000\">Monday 07 May 3:30pm: Oord et al.: Pixel-RNN and Pixel-CNN</span>\n** Location: EEBF 4302\n* <span style=\"color:#000000\">Monday 14 May 3:30pm: Isola et al.: Pix2Pix</span>\n** Location: EEBF 4302\n* <span style=\"color:#000000\">Monday 21 May 3:30pm: Girshick, Donahue, Darrell & Malik: R-CNN</span>\n** Location: EEBF 4302\n* <span style=\"color:#CAAAAA\">???: Redmon et al.: YOLO and YOLO9000</span>\n* <span style=\"color:#CAAAAA\">???: Khoreva et al.: Dense tracking/data augmentation</span>\n* <span style=\"color:#CAAAAA\">???m: Mancini et al.: Obstacle detection</span>\n* <span style=\"color:#CAAAAA\">???: Ilg/Fischer et al.: FlowNet/FlowNet 2.0</span>\n* <span style=\"color:#CAAAAA\">???: Pagnutti et al.: RGBD semantic segmentation with CNN + surface fitting</span>\n* <span style=\"color:#CAAAAA\">???: Zhu et al.: CycleGANs</span>\n* <span style=\"color:#CAAAAA\">???: Yang et al.: Full 3D reconstruction from single depth view</span>\n* <span style=\"color:#CAAAAA\">???: Li et al.: Fully Convolutional Instance-aware Semantic Segmentation</span>\n* <span style=\"color:#CAAAAA\">???: Kim et al.: Solving CRF with CNN (depth image)</span>\n* <span style=\"color:#CAAAAA\">???: Liu et al.: Attribute Grammar Scene Reconstruction</span>\n* <span style=\"color:#CAAAAA\">???: Tatarchenko et al.: Multi-view 3D models </span>\n* <span style=\"color:#CAAAAA\">???: H\u00e4ne et al.: Single-view voxel reconstruction</span>\n* <span style=\"color:#CAAAAA\">???: Garg et al.: Geometry+CNN unsupervised</span>\n* <span style=\"color:#CAAAAA\">???: Xie et al.: Deep3D</span>\n* <span style=\"color:#CAAAAA\">???: Liu et al.: Convolutional Neural Field CRFs</span>\n* <span style=\"color:#CAAAAA\">???: Hong et al.: Semantic segmentation for robot behaviour</span>\n* <span style=\"color:#CAAAAA\">???: Liao, Huang, Wang, Kodagoda, Yu & Liu: Fuse with laser</span>\n* <span style=\"color:#CAAAAA\">???: Mirowski et al.: Learning to navigate</span>\n* <span style=\"color:#CAAAAA\">???: Finn & Levine: Visual prediction for planning</span>\n* <span style=\"color:#CAAAAA\">???: Roy & Todorovic: Neural Regression Forest</span>\n* <span style=\"color:#CAAAAA\">???: Hoeim et al.: Photo pop-up </span>\n* <span style=\"color:#CAAAAA\">???: Heitz et al.: Cascaded Classification Models</span>\n* <span style=\"color:#CAAAAA\">???: Li et al.: Feedback-enabled Cascaded Classification Models</span>\n* <span style=\"color:#CAAAAA\">???: Li et al.: Depth & Normals - CRF/regression</span>\n* <span style=\"color:#CAAAAA\">???: Han et al.: Bayesian object-level reconstruction</span>\n* <span style=\"color:#CAAAAA\">???: Liu et al.: Depth from semantics</span>\n* <span style=\"color:#CAAAAA\">???: Wu et al.: Repetitive scene structure</span>\n* <span style=\"color:#CAAAAA\">???: He et al.: Haze removal</span>\n* <span style=\"color:#CAAAAA\">???: Hassner et al.: Example-based Depth</span>\n* <span style=\"color:#CAAAAA\">???: Wu et al.: Repetition-based Depth</span>\n\n</div>\n\n= Details =\n\n<div style=\"opacity:0.8\">\n\n== Foley & Maitlin Chapter 6 - Distance & Size Perception ==\n\nBecause 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...\n\nhttps://books.google.com.tr/books?id=jLBmCgAAQBAJ&printsec=frontcover\n\nGo to Chapter 6.\n\nIf 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. \n\nAnother thing to try that has worked for some is to log out of any google/gmail account before trying to access.\n\nIf nothing else works, email me.\n\n== Saxena, Min & Ng: Make3D ==\n\nhttp://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4531745\n\nThis 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.\n\n'''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.\n\nBut, here is an alternative link: http://www.cs.cornell.edu/~asaxena/reconstruction3d/saxena_make3d_learning3dstructure.pdf\n\nThere 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\n\nAfter that other datasets started being used also.\n\nSuperpixels are used in the study. Here is a quick intro to them: http://ttic.uchicago.edu/~xren/research/superpixel/\n\nMRFs 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.\n\n== Michels, Saxena & Ng: High speed obstacle avoidance ==\n\nhttp://dl.acm.org/citation.cfm?id=1102426\n\nHere the same authors focus on a related problem, that of determining open spaces for guiding a vehicle, again using machine learning techniques.\n\nThis version of the paper might be of higher quality (thanks to Hossein for finding):\n\nhttp://ai.stanford.edu/~asaxena/rccar/ICML_ObstacleAvoidance.pdf\n\n== Karsch, Liu & Kang: Depth Transfer ==\n\nHere is the target paper:\nhttp://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=5551153\n\nFor those who are not on campus, a temporary link:\nhttp://web.itu.edu.tr/djduff/Share/KarschEtAl2014.pdf\n\nThis 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.\n\nHere is a paper describing \"SIFTFlow\" (if you have the time to go deeper):\nhttp://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=6787109\n\nFree version:\nhttp://people.csail.mit.edu/celiu/SIFTflow/\n\nOr a shorter conference version available off-campus:\nhttp://people.csail.mit.edu/celiu/ECCV2008/\n\n== LeCun, Bengio & Hinton: Deep Learning Review ==\n\nhttp://www.nature.com/nature/journal/v521/n7553/full/nature14539.html?foxtrotcallback=true\n\nAlternative links: http://pages.cs.wisc.edu/~dyer/cs540/handouts/deep-learning-nature2015.pdf https://www.researchgate.net/publication/277411157_Deep_Learning\n\nA whirlwind compressed intro to deep learning and its parts.\n\nFor a more gentle introduction to deep learning: http://cs231n.stanford.edu/\n\nOr you can find lots of gentle short intros: https://www.google.com.tr/search?q=intro+to+deep+learning\n\n== Rumelhart, Hinton & Williams: Backpropagation ==\n\nAn early paper introducing backpropagation, the main way we train neural networks nowadays: http://www.nature.com/articles/323533a0\n\nAlternative link: http://www.cs.toronto.edu/~hinton/absps/naturebp.pdf\n\nThese 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 ).\n\n== LeCun, Bottou, Bengio & Haffner: CNNs ==\n\nThis 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.\n\nhttp://ieeexplore.ieee.org/abstract/document/726791/\n\nAlternative link: http://www.dengfanxin.cn/wp-content/uploads/2016/03/1998Lecun.pdf\n\nConvolutional 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\n\nOr any tutorial about CNNs. One good place to follow is: http://cs231n.stanford.edu/syllabus.html\n\n== Krizhevsky, Sutskever & Hinton: ImageNet/AlexNet ==\n\n<span style=\"color:#B00000\">We will not discuss this in the reading group.</span>\n\nhttps://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks.pdf\n\nHere is when convolutional neural networks and deep learning really showed what it could do - the problem of image recognition. \n\nBut 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.\n\n== Simonyan & Zisserman: VGG-16 ==\n\nhttp://arxiv.org/abs/1409.1556\n\nA relatively recent \"deep\" deep net with 16 layers for image recognition. Note: successful recent networks have one thousand layers.\n\nFeel free to take a look at AlexNet above to get an idea of the space of approaches.\n\n== Eigen, Puhrsch & Fergus: Depth map prediction ==\n\nhttps://www.cs.nyu.edu/~deigen/depth/\n\nFinally, we apply deep neural convolutional networks to the problem that we are interested in.\n\n== Luo et al.: Deep Learning + Stereo ==\n\nCombining deep learning and stereo.\n\nhttps://www.cs.toronto.edu/~urtasun/publications/luo_etal_cvpr16.pdf\n\n== Shelhamer, Long & Darrell: Fully Convolutional Segmentation ==\n\nhttp://arxiv.org/abs/1605.06211\n\nHere a related problem is solved, that of semantic segmentation, but this approach is applicable to our problem.\n\n== Giuisti et al.: Forest trails CNN ==\n\nhttp://ieeexplore.ieee.org/document/7358076/\n\nAlternative link: http://rpg.ifi.uzh.ch/docs/RAL16_Giusti.pdf\n\nSee also youtube: https://www.youtube.com/watch?v=umRdt3zGgpU\n\nHere we have a CNN-based update to the learn-to-navigate-from-images problem addressed by Saxena et al. above.\n\n== Loffe & Szegedy: Batch Normalization ==\n\nhttps://arxiv.org/abs/1502.03167\n\nA recent technique that has enabled powerful new methods and ultimately much deeper neural networks. Important stuff.\n\nApparently the following video from the Stanford CS231N course contains details of Batch Normalization (37:00 to 59:30)\n\nhttps://youtu.be/gYpoJMlgyXA?list=PLkt2uSq6rBVctENoVBg1TpCC7OQi31AlC\n\nBest,\nDamien\n\n== He, Zhang, Ren & Sun: ResNets ==\n\nhttps://arxiv.org/abs/1512.03385\n\nThis work and variations on it have been the basis of the 1000 layer recent neural networks. Important stuff.\n\nThe 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:\n\nhttp://arxiv.org/abs/1603.05027\n\n== Cao, Wu & Shen: Fully convolutional depth 1 ==\n\nhttp://arxiv.org/abs/1605.02305\n\nHere we start a series of recent papers that take different approaches using deep nets to depth from a single image.\n\nLet me know if you can find some media for this work (I could only find the paper itself).\n\n== Laina et al.: Fully convolutional depth 2 ==\n\nhttp://arxiv.org/abs/1606.00373\n\nHere we continue a series of recent papers that take different approaches using deep nets to depth from a single image.\n\n== Li, Klein & Yao: Fully convolutional depth 3 ==\n\nhttp://arxiv.org/abs/1607.00730\n\nHere we finalise a series of recent papers that take different approaches using deep nets to depth from a single image.\n\n== G\u00fcler et al. DenseReg ==\n\nhttps://arxiv.org/abs/1612.01202\n\nA key idea here is how to do regression using categorical prediction & an application of the Fully Convolutional Networks to regression problems.\n\n== Godard et al. Unsupervised/train from stereo ==\n\nhttps://arxiv.org/abs/1609.03677\n\nhttp://visual.cs.ucl.ac.uk/pubs/monoDepth/\n\nhttps://github.com/mrharicot/monodepth\n\n== Zhou et al.: SfMLearner ==\n\nSuper cool stuff.\n\nhttps://people.eecs.berkeley.edu/%7Etinghuiz/projects/SfMLearner/\n\nhttps://arxiv.org/abs/1704.07813\n\nA blog entry explaining the main ideas: http://bair.berkeley.edu/blog/2017/07/11/confluence-of-geometry-and-learning/\n\nThe code: https://github.com/tinghuiz/SfMLearner\n\n== Fangchang & Karaman: depth from a single image & SLAM ==\n\nhttp://www.mit.edu/~fcma/\n\nhttps://youtu.be/vNIIT_M7x7Y\n\nhttps://arxiv.org/pdf/1709.07492.pdf\n\nhttps://github.com/fangchangma/sparse-to-dense.git\n\n== Pizzoli et al.: REMODE ==\n\nIn 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).\n\n[1]M. Pizzoli, C. Forster, and D. Scaramuzza, \u201cREMODE: Probabilistic, monocular dense reconstruction in real time,\u201d in Robotics and Automation (ICRA), 2014 IEEE International Conference on, 2014, pp. 2609\u20132616. \n\nhttp://rpg.ifi.uzh.ch/docs/ICRA14_Pizzoli.pdf\n\nhttps://www.youtube.com/watch?v=QTKd5UWCG0Q\n\n== Yan et al.: Superpixel CNN/CRF ==\n\nhttp://ieeexplore.ieee.org/document/8105853/\n\n== Mnih et al.: Deep reinforcement learning ==\n\nA modern classic.\n\nhttp://arxiv.org/abs/1312.5602\n\nhttps://www.nature.com/articles/nature14236 (temporary link: http://web.itu.edu.tr/djduff/2018/nature14236.pdf )\n\nhttps://www.youtube.com/watch?v=iqXKQf2BOSE\n\nHere it is done in Keras: https://keon.io/deep-q-learning/\n\n== Dosovitskiy, Springenberg, Tatarchenko & Brox: Generating images ==\n\nhttps://arxiv.org/abs/1411.5928\n\nhttps://ieeexplore.ieee.org/document/7469347/media\n\nA non-adversarial approach to generating images.\n\nAlso see:\n\nhttps://www.youtube.com/watch?v=QCSW4isBDL0\n\nhttps://www.youtube.com/watch?v=LAfmJQK4UW0\n\n== Goodfellow et al.: Generative Adversarial Nets ==\n\nhttps://papers.nips.cc/paper/5423-generative-adversarial-nets\n\nAnother important recent development that we may make use of.\n\n== Oord et al.: Pixel-RNN & Pixel-CNN ==\n\nhttps://arxiv.org/abs/1601.06759\n\nProducing distributions over images. We have always intended to do something like this for depth images.\n\nhttp://arxiv.org/abs/1606.05328\n\nSome background on LSTMs:\n\nhttp://colah.github.io/posts/2015-08-Understanding-LSTMs/\n\n== Isola et al. Pix2Pix ==\n\nhttps://arxiv.org/abs/1611.07004 \n\nWe can use this too. And it's cool.\n\nThe project page:\n\nhttps://phillipi.github.io/pix2pix/\n\nAn online demo:\n\nhttps://affinelayer.com/pixsrv/\n\n== Girshick, Donahue, Darrell & Malik: R-CNN ==\n\nhttps://arxiv.org/abs/1311.2524\n\nWe 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.\n\n== Redmon et al.: YOLO and YOLO9000 ==\n\nhttps://arxiv.org/abs/1612.08242\n\nhttps://arxiv.org/abs/1506.02640\n\n== Khoreva et al.: Dense tracking/data augmentation ==\n\nAn attempt on the DAVIS dataset. The dataset is here: http://davischallenge.org/\n\nThis is interesting because of the data augmentation approach used.\n\nThe paper: http://arxiv.org/abs/1703.09554\n\nVideo: https://www.youtube.com/watch?v=QrsR5w-HR14\n\n== Mancini et al.: Obstacle detection ==\n\nhttps://arxiv.org/abs/1607.06349\n\n== Pagnutti et al. RGBD semantic segmentation with CNN + surface fitting ==\n\nhttp://ieeexplore.ieee.org/document/8120042/\n\nhttps://pdfs.semanticscholar.org/7716/9ee225157e77d1632e3bed54c70235b4abf0.pdf\n\n== Zhu et al.: CycleGANs ==\n\nhttps://arxiv.org/abs/1703.10593\n\nHere is a nice tutorial:\n\nhttps://hardikbansal.github.io/CycleGANBlog/\n\n== Ilg/Fischer et al.: FlowNet/FlowNet 2.0 ==\n\nFlowNet: \n\nhttps://www.youtube.com/watch?v=g-peWXaQnQc\n\nhttps://arxiv.org/abs/1504.06852\n\nFlowNet 2.0:\n\nhttps://www.youtube.com/watch?v=JSzUdVBmQP4\n\nhttps://arxiv.org/abs/1612.01925\n\n== Yang et al.: Full 3D reconstruction from single depth view ==\n\nhttps://arxiv.org/abs/1708.07969\n\n== Li et al. Fully Convolutional Instance-aware Semantic Segmentation ==\n\nhttps://arxiv.org/abs/1611.07709\n\n== Kim et al. Solving CRF with CNN (depth image) ==\n\nhttp://arxiv.org/abs/1603.06359\n\n\n== Liu et al. Attribute Grammar Scene Reconstruction ==\n\nhttp://ieeexplore.ieee.org/document/7889053/?source=tocalert&dld=Z21haWwuY29t\n\n== Tatarchenko et al. Multi-view 3D models ==\n\nhttps://arxiv.org/abs/1511.06702\n\nNot just inferring the depth image but also other views of it (related to the SfMLearner paper).\n\n== H\u00e4ne et al. Single-view voxel reconstruction ==\n\nBlog summary: http://bair.berkeley.edu/blog/2017/08/23/high-quality-3d-obj-reconstruction/\n\nVideo intro: https://www.youtube.com/watch?v=BjwhMDhbqAs\n\nFull paper: https://arxiv.org/abs/1704.00710\n\n== Garg et al. Geometry+CNN unsupervised ==\n\nWe will have already read Godard et al. and Zhou et al. but this is for completeness.\n\nhttp://arxiv.org/abs/1603.04992\n\n== Xie et al. Deep3D == \n\nWe will have already read Godard et al. and Zhou et al. but this is for completeness.\n\nhttps://arxiv.org/pdf/1604.03650\n\n== Liu et al. Convolutional Neural Field CRFs ==\n\nhttps://arxiv.org/abs/1411.6387\n\n== Hong et al.: Semantic segmentation for robot behaviour ==\n\nhttps://arxiv.org/abs/1802.00285\n\n== Liao, Huang, Wang, Kodagoda, Yu & Liu: Fuse with laser ==\n\nhttps://arxiv.org/abs/1611.02174\n\nHere we see an interesting depth-from-single-image sensor fusion with robotics applications.\n\n== Mirowski et al.: Learning to navigate ==\n\nhttps://arxiv.org/abs/1611.03673\n\n== Finn & Levine: Visual prediction for planning ==\n\nhttp://arxiv.org/abs/1610.00696\n\n== Hoeim et al. Photo pop-up ==\n\nA classic.\n\nhttp://repository.cmu.edu/cgi/viewcontent.cgi?article=1288&context=robotics\n\n== Roy & Todorovic: Neural Regression Forest ==\n\nhttps://www.cv-foundation.org/openaccess/content_cvpr_2016/app/S23-11.pdf\n\n== Heitz et al. Cascaded Classification Models ==\n\nOlder pre-CNN machine learning papers for depth estimation from a single image.\n\nhttp://papers.nips.cc/paper/3472-cascaded-classification-models-combining-models-for-holistic-scene-understanding.pdf\n\n== Li et al. Feedback-enabled Cascaded Classification Models ==\n\nOlder pre-CNN machine learning papers for depth estimation from a single image.\n\nhttps://arxiv.org/abs/1110.5102\n\n== Li et al. Depth & Normals - CRF/regression ==\n\nhttps://www.cv-foundation.org/openaccess/content_cvpr_2015/app/1B_001.pdf\n\n== Han et al. Bayesian object-level reconstruction ==\n\nhttp://escholarship.org/uc/item/9tk6935x.pdf\n\n== Liu et al. Depth from semantics ==\n\nOlder pre-CNN machine learning papers for depth estimation from a single image.\n\nhttp://ai.stanford.edu/people/koller/Papers/Liu+al:CVPR10.pdf\n\n== Wu et al. Repetitive scene structure ==\n\nOlder pre-CNN machine learning papers for depth estimation from a single image.\n\nhttp://www.academia.edu/download/30713855/WuCVPR11.pdf\n\n== He et al. Haze removal ==\n\nMight be interesting because of use of single-image cues.\n\nhttp://mmlab.ie.cuhk.edu.hk/2009/dehaze_cvpr2009.pdf\n\n== Hassner et al. Example-based Depth ==\n\nSeems like an older version of the SIFTFlow based one of Karsch.\n\nhttps://www.researchgate.net/publication/4245893_Example_Based_3D_Reconstruction_from_Single_2D_Images\n\n== Wu et al. Repetition-based Depth ==\n\n\"Repetition-based dense single-view reconstruction\"\n\nhttp://www.academia.edu/download/30713855/WuCVPR11.pdf\n\n</div>\n\n= Additional Resources =\n\nEBook: 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\n\nOnline course with slides videos and assignments: http://cs231n.stanford.edu/ CS231n: Convolutional Neural Networks for Visual Recognition (Stanford)\n\nMy NN/Keras bootcamp slides: http://files.djduff.net/nn.zip\n\nFoley & Maitlin's book: https://books.google.com.tr/books?id=jLBmCgAAQBAJ&printsec=frontcover"
}
]
},
"127": {
"pageid": 127,
"ns": 0,
"title": "Slot Game",
"revisions": [
{
"contentformat": "text/x-wiki",
"contentmodel": "wikitext",
"*": "online casino slots<br>[http://bom777casino.com online casino]<br>slot online<br>[http://bom777casino.com best online casino]<br>online casino games"
}
]
}
}
}
}