Friday, 2 February, 2018. Written later

On our last day in San Francisco we slept in, not having to get up early to go to my conference. We dozed until 08:30, then got up, showered, finished the last remnants of the breakfast cereals, packed our bags for the flight home, and checked out of the hotel, leaving our bags with the concierge while we went out for a last day of sightseeing before our evening flight home.

I realised we wouldn’t be able to take the hot sauce I’d bought for my mum onto the plane in cabin baggage, and we didn’t want to check luggage in, so we decided to mail the bottles back home. I searched online to find nearby post offices and found one that was supposed to be on the block that Macy’s was on, south of Union Square. I figured it might take some the post office, so went there while M. detoured via Blue Bottle to get a morning coffee. We arranged to meet at the front door of Macy’s. As it turned out, I couldn’t find the post office from the street, and so decided to go into Macy’s to use the toilets. As I was looking for signs indicating where the toilets were, I saw a sign saying that there was a post office in the basement! So I went downstairs and found it, a small office with some counters and not much else. I’d been hoping to find packaging materials, but there was a sign saying they didn’t have any. However, I found some prepaid mailing boxes, one size of which was perfect, and just needed a little bit of padding improvised by scrunching up the paper bag that I’d been carrying the bottles in. So that was fortunate. It cost $36 in postage, more than the hot sauce was worth, but it was good not to have to check bags in and then spend time waiting in Sydney before getting home.

Mara’s Italian Pastry

We met at the designated spot and then walked up to North Beach through Chinatown, to Mara’s Italian Pastry, for an Italian pastry each and some water. I felt like fizzy water, and M. asked the old Italian guy serving if they had any, and he said, “It’s Italian, of course we have fizzy water!” while I located the bottles of San Pellegrino in the fridge. M. had had an apricot danish-like pastry yesterday and said it was delicious and soft, so I tried the cherry version, while M., who had thought it was raspberry and wanted that, elected to try a raspberry and cream cheese ring instead. They were nice, but a bit firm and not as soft as M. had raved about, which was a bit of a shame. While we sat, the proprietor sat at the next table with another customer and chatted with him, getting up every now and then to serve someone else who came into the shop.

Read more: the Cartoon Art Museum, San Francisco Maritime National Historical Park, Golden Boy Pizza, and flying home

Thursday, 1 February, 2018. 16:59

On the BART train again, ready to head back into San Francisco after another interesting day at the conference. We got up and had cereal as usual this morning before I left. M.’s plan was to go to the International Museum of Art, just a block down Market Street from our hotel, and then head over to Cow Hollow, either by walking, or catching a bus.

I arrived at the conference and checked the schedule boards in front of the rooms, to notice some changes to the program I’d organised on the online schedule planner. The first talk I wanted to attend at the Human Vision & Electronic Imaging (HVEI) conference had been moved 20 minutes earlier, overlapping the end of the very first talk I’d planned to go to, which was about image sensors for automotive cameras. So I went to the first half of that and then had to leave partway through to make sure I didn’t miss any of the HVEI talks. The sensor talk was by Boyd Fowler, who is a good speaker, but his subject is always very technical details of sensor chips, often down to the transistor and capacitor level, and this talk was no exception. Which is fine for people into that stuff, but it’s not my cup of tea, so I didn’t mind leaving a bit early.

As it turned out, the first HVEI talk was given by the same woman who gave the talk I reacted poorly to last year. As last year, she spent the first part of the talk building up how wonderful she was. She said that not only was she a cognition scientist, but she also founded a theatre production company and has written a bunch of plays and produced several short films, as she showed a slide with shots of some of her productions, oh and she also writes music too. Then she talked about human/robot interaction and showed a slide with two images of women having dinner with and playing chess with a humanoid robot. She said she deliberately chose pictures with women and not men because wouldn’t it be wonderful if in the future there were just women and robots and no men, ha ha.

Read more: talks about human vision and art, meeting Tim Jenison, champagne and cake, and lots of garlic for dinner

Wednesday, 31 January, 2018. 17:36

On the train ride home again from the conference. I woke up early again, got up and had breakfast, then left to catch the train. M.’s plan for the day was to catch the ferry to Sausalito.

Today instead of walking along the road to the conference, I walked out to the pedestrian path that runs along the shore of the bay. This is a much nicer walk, especially in the morning light, as there are hundreds of shorebirds and an expanse of misty water. I saw ducks (mallards and possibly another species), Pacific gulls, curlews, willets, grebes, and a huge white egret.

Birds on the Bay in the morning

The talks today began with a keynote by Marc Levoy of Google, who talked about pushing the boundaries of low light imaging with phone cameras. Phones don’t have great apertures or zoom capability or sensor sizes, but they do have good low-noise sensors. He showed how you can use this to produce images in really low light conditions by shooting bursts of underexposed images, aligning them, and stacking them to reduce noise. He uses a cool trick to remove hot pixels by acquiring images during the autofocus sweep, which results in everything being blurry… except the hot pixels, so they are easily detected. Alignment is done on very large image patches, hundreds of pixels square, by brute force cross correlation which can be calculated in real time. This allows images to be stacked to produce clear real time video at about 20 frames per second, in moonlight light levels. At even lower light levels, his app can produce usable photos in light too dark to even walk in, around 0.1 lux.

Read more: more birdspotting, ice cream sandwich for lunch, the guy who invented augmented reality, Thai for dinner, and freshly baked cookies in a bar

Tuesday 30 January, 2018. 17:32

I’m writing on the BART train from Millbrae back to San Francisco. I had a better sleep last night, but woke up before the alarm went off at 07:00. We got up and I had breakfast quickly and dashed out so I could catch the same train I got yesterday. There were a few more people on it today, but there were still plenty of seats.

After walking to the conference hotel, I checked in at the AV prep room, where they test to make sure your presentation laptop works with their projector system. Everything was fine, and I also asked for any tips in chairing a session, since I had to do that job for the first time this afternoon. The guy said there would be AV people running around the conference rooms to help if anyone had any problems.

Then it was off to the first session of the day. Today I spent the whole day in the Photography, Mobile, and Immersive Imaging conference, since I had my own talk to give before lunch and then a session to chair in the afternoon. First up Henry Dietz gave a talk about the development of multicamera systems at the University of Kentucky, showing off several systems they’ve developed over the years and presenting lessons learnt on how to design and engineer systems made of multiple cameras. The most crucial thing he said is to recognise that cameras these days are actually computers with sensors, not boxes for exposing film, and we need to start taking more advantage of the programmability in order to unlock their potential. I always like Henry’s talks as they are casual and fun and show off really cool stuff that his research group are doing by hacking cameras.

Read more: my presentation at the conference, enormous Chinese lunch, meeting friends for dinner

Monday 29 January, 2018. 12:53

I am sitting in the Juban Yakiniku House in Burlingame, having lunch during the conference lunch break. I had another terrible sleep, lying awake much of the night, and did not feel like getting up when the alarm went off at 07:00. But I bounded out of bed and checked the BART timetable as I gulped down a mixture of Special K and bran cereal for breakfast. Trains left for Millbrae at 07:21 and 07:36. I hurried so I could make the first one, and raced out to leave M. alone for her first day of solitary sightseeing. I just made the train by a minute and settled in for the ride to Millbrae.

The train was largely empty at this time, which was good, as I had a double seat to myself. I went through the printout of my talk to help cement in in my brain, although I’m not giving it until tomorrow. Once at Millbrae, I set out for the walk to the Hyatt Regency Hotel where the conference is on. Checking the time at both ends, it came in at 28 minutes of walking. The morning was grey, but not especially cold, and I removed my gloves part way along as I warmed up from the exercise.

I arrived at the hotel at 08:20, leaving plenty of time to register and collect my conference badge before the first talk I planned to attend at 08:50. Even before entering, I saw Nicolas outside, dealing with his transport. Then inside I ran into Kristyn, who apologised for not answering my email, saying she was halfway through replying. Her reply arrived a few minutes into the first conference session, saying she was keen to meet up with M. again, but had commitments with her in-laws while here in the Bay Area.

The first talk session was a joint session between Automotive Imaging and Image Quality & Systems Performance, chaired by Stuart. And the very first talk was given by Robin Jenkin, who I met briefly outside before he dashed off saying he needed to get ready to present. He spoke about the task of measuring image quality of automotive cameras, and how it was very different from measuring the image quality of normal photographic cameras, mainly because photography quality standards all factor in the influence of the human visual system, which is absent in an automotive context. Also the colour filter arrays of car cameras tend to be very different from the RGB Bayer filter for human-oriented photography, typically car cameras use red-white-white-white, or red-white-white-blue. The output of a car camera goes into a neural network designed to produce a decision on what the car should do, not to produce an image for humans to look at, so completely different criteria need to be used to judge the quality of the camera.

Read more: more conference stuff, yakiniku lunch, Mexican dinner

Sunday 28 January, 2018. 12:52

We are sitting in Noe Bagel in Noe Valley, having a lunch break of bagels. We slept late this morning, having got to sleep fairly late, after 23:00 last night. I got a decent amount of sleep, but we were woke. around 02:00 by blaring sirens on the street outside that lasted several minutes. We got up after 09:00 and had breakfast and prepared to leave slowly, managing to get out close to 10:00.

First stop was Blue Bottle again for M.’s coffee. We sat in this time so I could type up some of yesterday’s diary. While we were there, we were sitting right in front of the counter where they made fancy “siphon coffee“. Some guy had ordered one and we watched while a guy spent what felt like about 15 minutes making it, involving multiple bits of glassware that looked like they belonged in a chemistry lab. First he boiled water in a spherical flask over a heat lamp. He used the steam rising from the water to clean and polish the inside of a glass funnel like thing. Once it was clean to his satisfaction, he inserted a filter, drawing it down until it sealed by pulling a chain through the narrow part of the funnel. Then he attached the funnel into the top of the boiling water flask with a rubber seal. When he did this, the funnel sucked the hot water up into it. Then he prepared a tray with a glass of iced water and two spoons in it, a shot glass of water, and another tall glass, plus a small white ceramic bowl. He grabbed some coffee beans and ground them in a machine, putting the result into a metal cup. Then he stirred the hot water in the funnel with an icy cold spoon and measured the temperature of the water by dipping a thermometer into it. After confirming the temperature, he poured the ground beans in. He let that sit for a while, timing it with a digital timer clock, and put some boiling water into the tall glass on the tray. When the timer went off, he removed the flask from the heat and stirred the mixture of hot water and ground beans so that it drained down into the flask again through the filter in a swirl, leaving a conical mound of dried bean grounds on top. He removed the funnel and tipped the beans into the ceramic cup on the tray, then poured out the hot water in the glass. He poured some of the coffee from the flask into the shot glass, then he took the cup of grounds and sniffed them, deeply several times, then he tasted the shot glass of coffee. Finally he poured the coffee from the flask into the heated tall glass, and took the tray out to the customer, complete with the glass of cold water and the cup of leftover ground beans. We were boggled at how long it took and how complicated it was. We figured if ten people came in and ordered this sort of coffee all at once, it’d take an hour for them to make them all.

After the coffee, we went down to Powell Street BART station to add some credit onto our Clipper cards. The machine refused my Visa card for some reason, so we used cash. Then we went back up to the street to catch an F bus to Castro. The bus took a while to arrive. I was hoping for one of the historic streetcars, but a bus arrived first so we got on that. It’s a fair distance to Castro so I’m glad we didn’t decide to walk all the way.

At Castro we started walking south down Castro Street towards Noe Valley, which was our first real goal for the day. One of the first shops we passed was a place selling cookies, called Hot Cookie, which looked very tempting. All the different types of cookies had suggestive names. We decided to try a Walnut Woody, which turned out to also have chocolate chips in it, which were all molten as the cookie was still warm. The guy weighed it to determine the price. We shared it as we walked down the street outside.

We crossed to the sunny side to look in a shop, but I had to step outside after starting to sneeze uncontrollably, possibly from the incense they were burning inside, as the sneezing stopped once I was outside. At the end of the shops we crossed back to the shady side for the hike up the steep hill. We needed to get to the other side of the hill to reach Noe Valley. From the top we had some views across various parts of the city, though houses blocked most directions. And as we descended into Noe Valley we noticed that Twin Peaks was just to the west, towering over the neighbourhood.

Florist in Noe Valley

Read more: Exploring Noe Valley, ice cream in Haight-Ashbury, jazz in Lower Haight, and burgers South of Market