Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Deep image prior 'learns' on just one image

dmitryulyanov.github.io

Post reply on HN
Print view View original

181–190 of 235 posts

Re: Deep image prior 'learns' on just one image

#181

Post by jon_richards »

stanfordkid wrote: post #110
cs702 wrote: post #21

Wow: "In this work, we show that, contrary to expectations, a great deal of image statistics are captured by the structure of a convolutional image generator rather than by any learned capability. This is particularly true for the statistics required to solve various image restoration problems, where the image prior is required to integrate information lost in the degradation processes. To show this, we apply untrain…

So are they saying that the topology of a deep-net is intrinsic to "reality" ... somewhat analogous to something like the Fibonacci ratio for organic forms?

Maybe more like a fourier transform.

Re: Deep image prior 'learns' on just one image

#182

Post by rsp1984 »

amelius wrote: post #134
rsp1984 wrote: post #131

Earlier quoted context omitted.

In the examples they show an image of a woman next to a table. The table is covered with a table cloth. In the corrupted image the table cloth looks like all noise to the human eye, however in the reconstructed image you see that the method is able to recover most of the pattern in the cloth.

That's something different: a demonstration of an "inpainting" problem. The black pixels were specifically marked as corrupt. Also, the pattern on the cloth is much less granular than noise.

Ah, correct. Since the image was in grey I thought is was added noise, not missing pixels. Thanks for pointing that out.

Re: Deep image prior 'learns' on just one image

#183

Post by eanzenberg »

jacquesm wrote: post #142
eanzenberg wrote: post #98

Earlier quoted context omitted.

Seems like this view gets told every once in a while by someone who clearly hasn't been around any 0-2 year olds.

Your comment is hilariously wrong. Please do not make assumptions like this, you're typically going to be embarrassed.

Please show me a child that recognizes a new object after one look.

Re: Deep image prior 'learns' on just one image

#184

Post by ConfucianNardin »

mjburgess wrote: post #175
ConfucianNardin wrote: post #173

Earlier quoted context omitted.

No, the issue is with the very premise, then. It is not clear that consciousness necessarily only emerges from a neurochemical process. What was basically said was "We do not know for certain that it is this neurochemical process that accounts for consciousness. Isn't it reasonable to suppose that we might encounter things that actually are conscious, but whose consciousness is not accounted for by the same property…

Scientific claims are not necessities. I'm not saying I can prove consciousness is a biological process only that its overwhelming reasonable to suppose so. Emergence is a result of causal interactions between parts of a system being different than the internal causal interaction within one part. It doesnt mean "complexity" and it really has nothing to do with a machine. The oscillating electric field acquires no new…

I do not find it bizarre to pose that it could be possible to use another set of primitives (than biochemical ones) to create something analogous to the higher level structure in a human brain that produces thoughts.

Re: Deep image prior 'learns' on just one image

#185

Post by fjsolwmv »

lern_too_spel wrote: post #95
goialoq wrote: post #90

Earlier quoted context omitted.

I don't understand their editorializing > contrary to expectations, ingore the weasel words > a great deal of image statistics are captured by the structure of a convolutional image generator rather than by any learned capability. > To show this, we apply untrained ConvNets to the solution of several such problems. Instead of following the common paradigm of training a ConvNet on a large dataset of example images, we…

The point is that a fully connected network would not get you the same results. I.e. the convolutional structure allows it to learn interesting things from a single example because by choosing a convolutional architecture, you are implicitly imposing a prior.

How is a network structure different from a fully connected network with 0 weights?

Re: Deep image prior 'learns' on just one image

#186

Post by fjsolwmv »

ska wrote: post #171
vbuwivbiu wrote: post #48

Remember fractal image compression ? https://youtu.be/AjdogjBxfco?t=260

This can also be looked at as the original source of patch based denoising, etc. In the end it's about capturing the scaling properties and self similarity of natural images. This is also, for example, why wavelets were so effective as a basis. David Mumford particularly did some great work on this sort of thing a couple of decades ago, along with many others. I hope when people are rushing around trying to apply con…

The benefit of CNNs is like the benefits of SVMs -- they generalize all the great old techniques so you don't have to understand them all, you just throw more CPU at the optimization problem.

Re: Deep image prior 'learns' on just one image

#187

Post by fjsolwmv »

rahimnathwani wrote: post #87
jacquesm wrote: post #77

Earlier quoted context omitted.

There is some pretty strong evidence for this: all the toddlers in the world. You only need to show them something once and they'll immediately be able to recognize more examples of the same thing from different angles and even when it is partially hidden. All they have to guide them is the structure of their brains, not the quantity of data they have been exposed.

"All they have to guide them is the structure of their brains, not the quantity of data they have been exposed." A typical toddler (say 12 months' old) has spent 4000-5000 hours with open eyes. Even if you assume a low frame rate (10fps), resolution (1080p), and a 1000:1 compression ratio, that's still 1TB of training data.

Consider language acquisition:

https://en.m.wikipedia.org/wiki/Poverty_of_the_stimulus

Re: Deep image prior 'learns' on just one image

#188

Post by syrak »

flyingspork wrote: post #33
JumpCrisscross wrote: post #24

Can someone break this down for this layman?

Not an expert so take this with a grain of salt; I could be misinterpreting the paper. It seems that the current accepted method is to train a network with distorted images as the input and the correct undistorted images as the targets. Then after training you can feed a new distorted image into the trained network and get the estimated "fixed" image. However this team actually uses the distorted image as both the in…

> So if they were to let the training go on for too long the network will produce an exact copy of the distorted input image.

It won't. The objective (min (E(x, x0) + R(x)) they are trying to optimize (over output images x) amounts to a combination of:

- The output image x should "look like" the input image x0, this is the error term E(x, x0);

- The output image x should be "regular", this is the regularization term R(x).

The latter term prevents overfitting: R should be chosen such that noisy images for example are considered irregular (high R(x)).

Re: Deep image prior 'learns' on just one image

#189

Post by thefalcon »

theoh wrote: post #36
oh_sigh wrote: post #26

Earlier quoted context omitted.

Neural networks are powerful without even training them. By merely designing the structure of them you are creating something.

Truly, an occasion for this koan: Sussman attains enlightenment In the days when Sussman was a novice, Minsky once came to him as he sat hacking at the PDP-6. “What are you doing?”, asked Minsky. “I am training a randomly wired neural net to play Tic-Tac-Toe” Sussman replied. “Why is the net wired randomly?”, asked Minsky. “I do not want it to have any preconceptions of how to play”, Sussman said. Minsky then shut hi…

Okay, I admit it, I'm not enlightened. Will somebody please Explain Like I'm 5?

Re: Deep image prior 'learns' on just one image

#190

Post by kkylin »

cs702 wrote: post #21

Wow: "In this work, we show that, contrary to expectations, a great deal of image statistics are captured by the structure of a convolutional image generator rather than by any learned capability. This is particularly true for the statistics required to solve various image restoration problems, where the image prior is required to integrate information lost in the degradation processes. To show this, we apply untrain…

Probably more than structure (which I took to mean connectivity) -- there are a lot of things that affect (biological) neuronal computation in addition to connectivity, such as synaptic weights, distribution & types of receptors for different neurotransmitters, and the (sometimes very different) timescales associated with different neurotransmitters. Not easy things to measure.
Post reply on HN