Assignment 1 - Bad Photos

Traffic Lights at the Corner of Matadero Ave and El Camino Real. Photo by Andrew Adams


Cameras in automatic mode try to take care of the technical aspect of photography for you, and just let you frame the scene and shoot. They will attempt to make sure the scene is not too dark, not too bright, not blurry, and so on. These goals don't always match up with the photographer's intent. For this first assignment, you'll be playing with the settings on your camera to produce photos you can't easily produce in automatic mode. You'll be taking photos that your camera might consider "bad" in some technical way, but are nonetheless interesting to look at, and even artistic.

The assignments in this course are adapted from Stanford CS 178.  In the Google version of the course, doing the assignments is optional.  If you choose to do them, you'll use this assignment to set up the Google Photos / Google+ workflow that you'll use throughout the course - to share your work with your fellow enrollees and get feedback from them.


  1. Take some photos that meet the requirements listed below.

    Each photograph you upload should meet at least one of the requirements below, and you should make sure to cover every requirement with at least one photo. You should upload at least five photographs for the assignment, but no more than ten. While a photo can fulfill more than one requirement, you should select exactly one requirement for each photo, and indicate in the photo's caption which requirement you are fulfilling with that photo. Note that these general rules apply to this assignment and all subsequent photo assignments in this course.

    Now, for Assignment 1 specifically: remember, even though the photos you take are supposed to be technically "bad" in some sense, we want you to create things that are interesting to look at. The "badness" should be intentional for an artistic purpose.

    • Requirement 1: At least one photo must be poorly exposed. That is, most of the image should be either very close to black (underexposed) or close to flat white due to oversaturation (overexposed).
    • Requirement 2: The main subject of at least one photograph should be motion-blurred, either due to movement of the subject or movement of the camera.
    • Requirement 3: In at least one photo, nothing at all should be in focus. It's hard to take a good looking photo where nothing is in focus - be creative!
    • Requirement 4: You should use the wrong white balance setting for some intentional effect in at least one photo. The white balance setting on a camera tells the camera what color it should expect the scene illumination to be, for example daylight (which is bluish) or indoor incandescent (which is reddish). Most cameras have an "auto white balance (AWB)" setting, and manual settings for these other illumination types. You don't need to understand white balance in detail to fulfill this requirement. It's a subject we'll cover in detail later in the course.
    • Requirement 5: At least one photo should be poorly composed. Scan through Chapter 12 of the London textbook if you've bought it, and search on Google for "rules of composition". Read about them, and then intentionally break a few. If you're not sure what to do you may want to consider an exactly centered subject that produces an oddly symmetric photo, a confusion between the subject and a background object, or a horizon that isn't level. We will cover photographic composition in lecture in 2 weeks.

  2. Upload your assignment to Google Photos as an album, organize the album however you like, and add a caption to each photo describing which requirement it satisfies, as described above.  See the FAQ question below about upload resolution.

  3. Post all the photos in your album to the course's Google+ community. To begin a post, click on "What's new with you?" and give your post a title, like "Submission of assignment #1 (Bad Photos) by <my name>" (substituting your name). Then click on Attach: Photos, select Your Google Photos, and find and select your photos.  You can select multiple photos using shift-click, then re-order them and caption them. Captions you already inserted in Google Photos will appear, which you can modify or replace.  Your modified captions will appear only in your Google+ post; your Google Photos remain unmodified. Finally, select a category (Assignment #1) and click Share.

    After your post appears in the community, you will have the ability to delete it, edit it, or add comments to it, if you have anything to say about the album as a whole.  You may also add comments to any photo in the post, but it's better to use captions - on the photos in Google Photos or while posting them - in order to keep your description of each photo separate from comments others will make about that photo while reviewing it (see below).  BTW, any comments you or others make will appear in the Google+ post, but not in your Google Photos album.

     [Steps 2 and 3 were designed for the Google version of this course, and are irrelevant to people who are following these assignments on their own based on these web pages.]

Example Solution

Here is an example solution to this assignment to let you know what I expect. This example solution was created using Picasa Web, so the user interface is archaic. Again, your comments should be entered as captions in Google Photos before posting to photos to the Google+ community, not as comments in the Google+ posting (despite the way it's done in the sample solution). For additional inspiration, you may want to peruse the best Stanford CS 178 photographs (as selected by the TA's and instructor) from 2009, 2010, 2011, 20122013, and 2014. However, you do not need to meet the assignment requirements in the same way as these solutions do - in fact, you should challenge yourself to come up with creative, unique compositions. There are plenty of possibilities out there. Surprise me (and your peers)!

Also, try to submit photographs taken by you specifically for this course. Do not reuse old photographs that you may have laying around, even if they perfectly meet one of the requirements.

Peer-to-Peer Commenting on Photos

After the assignment deadline, meaning on Monday, March 28, go to the Google+ Community, select the "Assignment #1" category in the left-hand sidebar, find 3 or more submissions that have fewer than 3 reviewers for them, open those posts, and add some comments on its photos.  For each photo try to use the following schema:  First, did the photographer satisfy the requirement he/she claims with that photograph? Second, was there something you particularly liked about the photograph?  Third, is there something you thought the photographer could have done differently, and could therefore work on in the future?  You don't need to comment on every photo, but try to comment on most of them. Keep your comments brief but constructive.  Be Googley!

Finally and importantly, back out of slideshow mode and make a comment on the entire post - to the effect, "Andrew Adams has reviewed this album", so that others know without having to open the post that it has been reviewed by 1 more person. Of course, if you find a particularly awesome picture in a post, feel free to expand your comment: "Awesome drone picture of the Golden Gate Bridge!"

Try to finish this commenting phase by the end of the week, i.e. by EOD on Friday, April 1, while the assignment is still fresh in everyone's mind. 

Practice Problems

To help you test your knowledge of the lecture material, I am giving out practice questions with each assignment. Note: answers have now been marked in bold.

  • Problem 1a. True or False? If camera apertures were square instead of circular, then F numbers will be separated by factors of 2 instead of sqrt(2).
  • Problem 1b. True or False? In aperture priority mode, if you lower the F-number, the image gets brighter.
  • Problem 1c. True or False? In-lens leaf shutters expose the center of the field of view longer than the corners.

  • Problem 2. Which of the following statements are true about pinhole photographs? Assume the surface on which the image is formed is flat. Circle all that apply.
    • (a) They are linear perspectives.
    • (b) They have infinite depth of field.
    • (c) Straight lines in the world can become curved in the photograph.
    • (d) They are upside down, but if you rotate them 180 degrees they look correct.
    • (e) They are upside down and if you rotate them they still read read backwards.
    • (g) They are dim, unless the scene is very bright or you use a very long exposure.
    • (h) They are in black and white.

  • Problem 3. What is the minimum number of vanishing points there can be in a linear perspective drawing of a cube? Assume the cube is an open wire frame (or it's made of glass, so you can see all 12 edges regardless of your viewpoint). Assume also that the cube measures 1 foot on a side, and you can't stand more than a few feet away from it. Circle the best answer.
    • (a) Zero
    • (b) One
    • (c) Two
    • (d) Three
    • (e) Infinitely many
  • There will be one vanishing point in a linear perspective image for every direction of light that is not parallel to the picture plane. No matter how you orient a cube, there will be at least one such direction, which goes "away from you". This direction will have a vanishing point.

  • Problem 4. You're taking a picture of an athlete at f/2.8, 1/200s, and ISO 200. You realize that you want twice as much depth of field and half as much motion blur. What new settings do you use, if you want to keep the same amount of exposure?
    • (a) f/4, 1/400s, ISO 200
    • (b) f/4, 1/400s, ISO 400
    • (c) f/4, 1/400s, ISO 800
    • (d) f/5.6, 1/400s, ISO 800
    • (e) f/5.6, 1/400s, ISO 1600
  • In order to double our depth of field we must use f/5.6, and in order to cut motion blur in half we must use 1/400s. The former decreases light by 4x and the latter decreases light by 2x, so we should increase our ISO by 8x to compensate, giving us answer (e).

Due Dates

Assignment Deadline: 11:59pm, Sunday, March 27, 2014
Commenting Deadline: EOD, Friday, April 1, 2014

Assignments are generally related to the material from the week before they are assigned, and are due on Sunday at midnight at the end of the week in which they're assigned. This leaves you free to start thinking about the next assignment during the next week's lectures, which are about a different topic.  It also ensures that you will get feedback on your photos from the peer-to-peer system.


Q: My Camera doesn't have manual focus. How can I force everything to be out of focus?

A: Even if your camera doesn't provide manual focus, you can usually force the focus to be anything you want. Just point at an object that's not in your scene but at the desired depth, press the shutter button halfway to autofocus on it, then with the button still half-pressed, swing around to aim at your scene and press the button fully. Does the exposure come out wrong when you do this? We'll talk in class about focus-lock and exposure-lock functions, which not all cameras have.

Q: At what resolution should I upload my photos?

A: Googlers don't have an upload limit that I am aware of, so you should set your Google Photos > Settings > Upload Size and Storage to Original (Full resolution at original quality). Doing this also ensures that Google Photo's Auto-Enhance functionality doesn't mess with your photographs.

Q: Are we allowed to edit our photos in Photoshop before uploading? How much editing is acceptable? Do we have to say what we did?

A: In general, most requirements won't necessitate any image manipulation in Photoshop, and for this first assignment, none of them do. However, some requirements for future assignments will specifically involve image editing, and even for those that don't, you're always welcome to process your images in Photoshop or any other program. Just keep in mind that if you do edit your photos in Photoshop, you should definitely state what you did in the comments. For example, you might say: "I used Photoshop to increase the saturation and darken the sky". Finally, keep in mind that Photoshop should not be used as a substitute for proper use of your camera or for faking a requirement. For example, for this assignment your underexposed or overexposed photo must be that way from the camera -- you're not supposed to just take a normal photo and use Photoshop to make it way darker or brighter.

Page authors: Marc Levoy, Andrew Adams, and Jesse Levinson, revised by Marc Levoy for the Google version of this course.