Mercurial > cortex

comparison org/literature-review.org @ 376:057d47fc4789

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
reviewing ullman's stuff.
author Robert McIntyre <rlm@mit.edu>
date Thu, 11 Apr 2013 05:40:23 +0000
parents 9c37a55e1cd2
children 80cd096682b2
comparison
equal deleted inserted replaced
375:cf6eea5d651b 376:057d47fc4789
40 partially occluded objects,” IEEE Trans. on Pattern Analysis and 40 partially occluded objects,” IEEE Trans. on Pattern Analysis and
41 Machine Intelligence, 19, 9 (1997), pp. 1043–48. 41 Machine Intelligence, 19, 9 (1997), pp. 1043–48.
42 42
43 - Zhang, Z., R. Deriche, O. Faugeras, Q.T. Luong, “A robust 43 - Zhang, Z., R. Deriche, O. Faugeras, Q.T. Luong, “A robust
44 technique for matching two uncalibrated images through the recovery 44 technique for matching two uncalibrated images through the recovery
45 of the unknown epipolar geometry,” Artificial In- telligence, 78, 45 of the unknown epipolar geometry,” Artificial Intelligence, 78,
46 (1995), pp. 87-119. 46 (1995), pp. 87-119.
47 47
48 48
49 49
50 50
51 51
52 * Alignment by Maximization of Mutual Information, Paul A. Viola 52 * Alignment by Maximization of Mutual Information, Paul A. Viola
53 53
54 PhD Thesis recommended by Winston. Describes a system that is able 54 PhD Thesis recommended by Winston. Describes a system that is able
55 to align a 3D computer model of an object with an image of that 55 to align a 3D computer model of an object with an image of that
56 object. 56 object.
61 which is worth reading, especially the part about entropy. 61 which is worth reading, especially the part about entropy.
62 62
63 - Differential entropy seems a bit odd -- you would think that it 63 - Differential entropy seems a bit odd -- you would think that it
64 should be the same as normal entropy for a discrete distrubition 64 should be the same as normal entropy for a discrete distrubition
65 embedded in continuous space. How do you measure the entropy of a 65 embedded in continuous space. How do you measure the entropy of a
66 half continuous, half discrete random variable? 66 half continuous, half discrete random variable? Perhaps the
67 problem is related to the delta function, and not the definition
68 of differential entropy?
67 69
68 - Expectation Maximation (Mixture of Gaussians cool stuff) 70 - Expectation Maximation (Mixture of Gaussians cool stuff)
69 (Dempster 1977) 71 (Dempster 1977)
70 72
71 - Good introduction to Parzen Window Density Estimation. Parzen 73 - Good introduction to Parzen Window Density Estimation. Parzen
72 density functions trade construction time for evaulation 74 density functions trade construction time for evaulation
73 time.(Pg. 41) 75 time.(Pg. 41) They are a way to transform a sample into a
76 distribution. They don't work very well in higher dimensions due
77 to the thinning of sample points.
78
79 - Calculating the entropy of a Markov Model (or state machine,
80 program, etc) seems like it would be very hard, since each trial
81 would not be independent of the other trials. Yet, there are many
82 common sense models that do need to have state to accurately model
83 the world.
84
85 - "... there is no direct procedure for evaluating entropy from a
86 sample. A common approach is to model the density from the sample,
87 and then estimate the entropy from the density."
88
89 - pg. 55 he says that infinity minus infinity is zero lol.
90
91 - great idea on pg 62 about using random samples from images to
92 speed up computation.
93
94 - practical way of terminating a random search: "A better idea is to
95 reduce the learning rate until the parameters have a reasonable
96 variance and then take the average parameters."
97
98 - p. 65 bullshit hack to make his parzen window estimates work.
99
100 - this alignment only works if the initial pose is not very far
101 off.
102
74 103
75 Occlusion? Seems a bit holistic. 104 Occlusion? Seems a bit holistic.
76 105
77
78 ** References 106 ** References
79 - "excellent" book on entropy (Cover & Thomas, 1991) 107 - "excellent" book on entropy (Cover & Thomas, 1991) [Elements of
80 108 Information Theory.]
109
110 - Canny, J. (1986). A Computational Approach to Edge Detection. IEEE
111 Transactions PAMI, PAMI-8(6):679{698
112
113 - Chin, R. and Dyer, C. (1986). Model-Based Recognition in Robot
114 Vision. Computing Surveys, 18:67-108.
115
116 - Grimson, W., Lozano-Perez, T., Wells, W., et al. (1994). An
117 Automatic Registration Method for Frameless Stereotaxy, Image
118 Guided Surgery, and Enhanced Realigy Visualization. In Proceedings
119 of the Computer Society Conference on Computer Vision and Pattern
120 Recognition, Seattle, WA. IEEE.
121
122 - Hill, D. L., Studholme, C., and Hawkes, D. J. (1994). Voxel
123 Similarity Measures for Auto-mated Image Registration. In
124 Proceedings of the Third Conference on Visualization in Biomedical
125 Computing, pages 205 { 216. SPIE.
126
127 - Kirkpatrick, S., Gelatt, C., and Vecch Optimization by Simulated
128 Annealing. Science, 220(4598):671-680.
129
130 - Jones, M. and Poggio, T. (1995). Model-based matching of line
131 drawings by linear combin-ations of prototypes. Proceedings of the
132 International Conference on Computer Vision
133
134 - Ljung, L. and Soderstrom, T. (1983). Theory and Practice of
135 Recursive Identi cation. MIT Press.
136
137 - Shannon, C. E. (1948). A mathematical theory of communication. Bell
138 Systems Technical Journal, 27:379-423 and 623-656.
139
140 - Shashua, A. (1992). Geometry and Photometry in 3D Visual
141 Recognition. PhD thesis, M.I.T Artificial Intelligence Laboratory,
142 AI-TR-1401.
143
144 - William H. Press, Brian P. Flannery, S. A. T. and Veterling,
145 W. T. (1992). Numerical Recipes in C: The Art of Scienti c
146 Computing. Cambridge University Press, Cambridge, England, second
147 edition edition.
148
149 * Semi-Automated Dialogue Act Classification for Situated Social Agents in Games, Deb Roy
150
151 Interesting attempt to learn "social scripts" related to resturant
152 behaviour. The authors do this by creating a game which implements a
153 virtual restruant, and recoding actual human players as they
154 interact with the game. The learn scripts from annotated
155 interactions and then use those scripts to label other
156 interactions. They don't get very good results, but their
157 methodology of creating a virtual world and recording
158 low-dimensional actions is interesting.
159
160 - Torque 2D/3D looks like an interesting game engine.
161
162
163 * Face Recognition by Humans: Nineteen Results all Computer Vision Researchers should know, Sinha
164
165 This is a summary of a lot of bio experiments on human face
166 recognition.
167
168 - They assert again that the internal gradients/structures of a face
169 are more important than the edges.
170
171 - It's amazing to me that it takes about 10 years after birth for a
172 human to get advanced adult-like face detection. They go through
173 feature based processing to a holistic based approach during this
174 time.
175
176 - Finally, color is a very important cue for identifying faces.
177
178 ** References
179 - A. Freire, K. Lee, and L. A. Symons, BThe face-inversion effect as
180 a deficit in the encoding of configural information: Direct
181 evidence,[ Perception, vol. 29, no. 2, pp. 159–170, 2000.
182 - M. B. Lewis, BThatcher’s children: Development and the Thatcher
183 illusion,[Perception, vol. 32, pp. 1415–21, 2003.
184 - E. McKone and N. Kanwisher, BDoes the human brain process objects
185 of expertise like faces? A review of the evidence,[ in From Monkey
186 Brain to Human Brain, S. Dehaene, J. R. Duhamel, M. Hauser, and
187 G. Rizzolatti, Eds. Cambridge, MA: MIT Press, 2005.
188
189
190
191
192 heee~eeyyyy kids, time to get eagle'd!!!!
193
194
195
196
197
198 * Ullman
199
200 Actual code reuse!
201
202 precision = fraction of retrieved instances that are relevant
203 (true-postives/(true-positives+false-positives))
204
205 recall = fraction of relevant instances that are retrieved
206 (true-positives/total-in-class)
207
208 cross-validation = train the model on two different sets to prevent
209 overfitting.
210
211
212
213
214
215 ** Getting around the dumb "fixed training set" methods
216
217 *** 2006 Learning to classify by ongoing feature selection
218
219 Brings in the most informative features of a class, based on
220 mutual information between that feature and all the examples
221 encountered so far. To bound the running time, he uses only a
222 fixed number of the most recent examples. He uses a replacement
223 strategy to tell whether a new feature is better than one of the
224 corrent features.
225
226 *** 2009 Learning model complexity in an online environment
227
228 Sort of like the heirichal baysean models of Tennanbaum, this
229 system makes the model more and more complicated as it gets more
230 and more training data. It does this by using two systems in
231 parallell and then whenever the more complex one seems to be
232 needed by the data, the less complex one is thrown out, and an
233 even more complex model is initialized in its place.
234
235 He uses a SVM with polynominal kernels of varying complexity. He
236 gets good perfoemance on a handwriting classfication using a large
237 range of training samples, since his model changes complexity
238 depending on the number of training samples. The simpler models do
239 better with few training points, and the more complex ones do
240 better with many training points.
241
242 The more complex models must be able to be initialized efficiently
243 from the less complex models which they replace!
244
245
246 ** Non Parametric Models
247
248 *** Visual features of intermediate complexity and their use in classification
249
250 *** The chains model for detecting parts by their context
251
252 Like the constelation method for rigid objects, but extended to
253 non-rigid objects as well.
254
255 Allows you to build a hand detector from a face detector. This is
256 usefull because hands might be only a few pixels, and very
257 ambiguous in an image, but if you are expecting them at the end of
258 an arm, then they become easier to find.
259
260