Map: https://www.tripoffice.com/maps
Yolo: https://www.ultralytics.com/yolo
The whole process was done on a home Mac without the use of any LLMs. It's based on traditional object detection technology.
Map: https://www.tripoffice.com/maps
Yolo: https://www.ultralytics.com/yolo
The whole process was done on a home Mac without the use of any LLMs. It's based on traditional object detection technology.
35 comments
https://youtube.com/watch?v=lVIGhYMwRgs
Hotel rooms suck when you need to use them for work. Typically there are massive beds and I travel alone if for work. There is no proper chair, no writing table at all or one too small, and the sockets tend to be in the wrong corner of the room.
If I was an entrepreneur outside of software looking for a gap, I might have invented a hotel chain for work stays. But I'm not, so here is the idea for you to get rich with it (so I can stay there one day).
I like the OP's idea of using ML models to gather intelligence from hotel photos. For years I took a photo of nearly every hotel room with my laptop on the desk so that I could go back and re-book the rooms that were suitable if there was a conference in the same city again in the future.
I think there's a simpler explanation: most people don't do work when they go to hotels, they do the work like you are discussing. Doesn't mean they are intentionally being hostile to remote workers.
One thing I've realized about the world is that a lot of people do things just because others are. "Momentum is a bitch." I will bet you very few people are thinking this way, at least very few that make decisions. And the ones that do probably think it is not worth the money. There's a ton of things where markets don't exist simply because the environment doesn't exist, so the people that can make the environment don't because there is no market. It's the whole "build it and they will come" thing. People are very risk adverse. People are hard to move. Would hotels benefit from this? Probably. I mean even not just considering nomads, most people work from their computers[0].
But it very easily could be one of those things where there's push because there's no market and there's no market because there's no push.
[0] The way people have been talking about working at CES has sounded silly. There was a LTT video where they mentioned how WiFi used to be better in some locations so those rooms were more desirable and the hotel's solution was to make it standard for everyone. They seemed to be suggesting that they brought down the quality rather than balanced.
And I think there’s a very good chance hotels would not benefit from this. Maybe in a tech center, but that’s a tiny fraction of hotels. Good office chairs are designed to be very adjustable, but they do tend to break when people twist one thing too far the wrong way because they don’t know how it works, and it would probably take staff 10 minutes just to figure out it was broken rather than just misadjusted. They’re also expensive as hell, and charging a guest $3000 because their luggage got caught on and tore up the mesh seat is probably not going to fly. Small higher-res monitors are also more expensive than huge TVs, and as or more delicate. The staff would spend more time than is probably worth it telling gran and gramps that they can’t use the “little television” like that. All of this stuff has to be handled with smoothness and grace 24/7 by a desk staff that don’t regularly use these items in their professional lives. You can’t just say “it’s a computer monitor gramps don’t use it” and hang up the phone. Many people also consider office equipment ugly, and how the room visually hits when you walk in is a huge consideration. Some weary overworked travel-worn office drone would probably want to jump out the window if they opened the door to their safe place of respite only to see a the better part of a corporate workstation looking back at them.
Designing experiences can be complicated and difficult, and that’s even more true because many of the most important aspects of it aren’t even consciously perceived by the intended audience. They all just fit organically unto a unified experience.
Working and sleeping in the same room is actually not that great for you most the time.
This was from their podcast 'WAN Show' a week or two back, specifically about hotels in Las Vegas.
So I too now care about a decent chair, desk and maybe even a tv I could turn into a second screen. Wifi can be there or not, I bring my own connectivity just to play it safe, this is now quite cheap. Bonus if the place is a couple time zones away from my office so I have my mornings or afternoons free.
I'm not a huge fan of AirBnB but it's been more reliable than hotels for a few of these factors: hotel TVs are locked-down and many won't accept an HDMI input, assuming there's a socket at all. Normally you're not offered (barring extravagant prices) more than just a bedroom, so the chance of table and chair being any good (or existing) are not so good), etc.
Otherwise, you’d only get a few hours per day in the evening of experiencing anything you couldn’t do at home, so what’d be the point of spending home rent +hotel +travel for the week?
It's fun. I occasionally get work done too.
I'm surprised ironing boards are so readily available.
I thought it was just a US thing when I was younger, but I've found it to be true even in other countries.
I took a look at business hotels in Japan. These are hotels explicitly designed for work travel and nothing more. Small rooms, bed, shower, but not much to actually work. And it actually makes sense. If you are on a work trip, why would you want to work in your hotel room? The whole point of a work trip is to visit a work place, that's where you are going to work, not your hotel room. In fact, from my limited experience of work travel, doing more work is the last thing I want to do when I am back at the hotel, it is often an exhausting day, and there is a good chance I have to get up early the next day, so the hotel room is for relaxation and sleep.
If you really want to work in your hotel room, or do anything other than using the bed and shower for that matter, you are probably better off with "apartment hotels" and short term rentals. If available, student residence rooms can be a minimal option for working and sleeping, that's what they are designed for. Note that there are also hotels with co-working spaces.
Maybe what you want, that is essentially a short-stay student room for grownups will happen one day, but I see many obstacles in making it a "get rich quick" investment. It may not be a great hotel for those who just want to sleep (or have other kind of fun). And if you want to eat in there, you will lack the amenities an appartement offers. And if you are not alone, a co-working space may be a better option.
The rooms in Clayton Bay Hotel in Hiroshima absolutely has a nice proper work desk and a work chair. So if anybody here is ever in Hiroshima Japan, you now know where to stay :)
Not sure if this applies to all room types though.
Disclaimer: I’m not related to that hotel in any way other than having stayed there one night some years ago.
There are usually plenty of "business" oriented hotels near airports, business parks, central business districts, convention centers, etc . (And that definitely reflects on the trip office map). Touristy areas have more tourist/traveller oriented amenities.
And I like having a king bed even if it's just me. (I do like having a desk and some sort of office chair though even that isn't really critical most of the time.)
I'm semi-retired now but I'm temporarily staying in a Marriott property (Springhill Suites) that does have a usable desk and office chair which is just fine for writing at for me--though people with very specific requirements probably wouldn't like it.
That's basically what wework is. I know you can't officially sleep there. I don't know what they would do if you slept in one of the 24/7 access plans though.
Also you are vastly overestimating the amount of "work" people do in hotel rooms that are not in tech.
(For an extended trip to a single location where I was only intermittently at a customer etc., maybe I'd consider asking for a co-working space but I never did and don't think I'd have gone to the trouble.)
Could also crop just the object detection regions of each image, run those cropped images through CLIP/SigLIP, then UMAP and HDBSCAN to view a 2 or 3 dimensional latent space clustering of office chair types.. might reveal some info as to what kinds of chairs exist in what geographical regions. Could use a VLM to auto-tag each cluster given a couple images from each one. Could run PCA on the CLIP embeddings and have some sliders for each principal component.. maybe the first is chair color or size or whatever
much data = much fun
I feel like they should be one database with object_type=car and object_type=firearm respectively. And then I can finally search by object_type=vacuum_cleaner and find out the wild-looking ball-shaped vacuum in that sci-fi movie whose name escapes me...
https://d3j17a2r8lnfte.cloudfront.net/mvh/2024/3/medium/bzSo...
My grandparents had one in the 70s, it always amazed me.
https://news.ycombinator.com/submitted?id=nomad86
I can understand why. My thought pattern was also like "oh YOLO, I know that, interesting to see some application for it, oh cool idea, upvote".
Side note; I love how YOLO, a deep learning based model, is now being referred to as traditional object detection. Template matching gang rise up.
I just looked at hotels I'm familiar with, and images with highlighted chairs were not limited to the guest rooms. Some are definitely from shared spaces at the hotel.
I trained an AI to recognize ergonomic chairs, but sometimes there were errors. For example, a chair in a hotel's SPA was always identified as an ergonomic chair. That's why we manually reviewed all 50k photos to verify them.
I edited that down to a proper summary of AI in general
I don't have an huge issue with short term rentals per si. They are an important niche for tourism when you take the whole family, whereas most (esp. city) hotels are not really appropriate for a family of 4 or 5.
OTOH, (esp. city) hotels are usually fine for the business trips they were designed to cater too.
This leaves us with “digital nomads.” Helping these find ways to put additional pressure on the housing market through short term rentals will only cause locals to get politicians to further restrict them.
That said, I would consider scraping, even with API access. In some ways the API access is both limited and binds you to their terms of service, and depending on the legalities in your jurisdiction, scraping could be more effective.
Do you hear that noise? It sounds like something is scraping
I want this to be a project that teaches me the ropes but since I need instant gratification, I'd like if the result also offered value to others.
The one thing I want to avoid is cleaning up data, since spreadsheets give me the ick.
https://github.com/IDEA-Research/Grounded-Segment-Anything
You can use that to take images and generate annotated segmented images/masks that you can then train a YOLO model on. I've done this for prototypes before and it's a very quick way of getting started as you can hand off the really annoying annotation work to a machine.
At TripOffice, we use simple and widely-used tools: Python, NextJS, and MySQL.
We are in several partner programs, but now we mainly work with HotelPlanner.
I use MapBox, but similar things can also be done on Google Maps.
This allows us to show our readers which specific products are in the photo they are looking at which is a service we offer to the manufacturers.
It is advertising, but like hyper-specific and relevant to the photo you’re looking at.
Moxy (Marriott) hotels tend not to have desks either.
Disclaimer: i am working for myself, i have no money, a 15 year old laptop and obsessions
Checkout as well darknet, which runs at really high fps on super cheap hardware
It can be hard to find a pub with a pool table these days!
https://thewirednomad.com is better suited because they are all non-hotels and have verified internet speeds.
Previously posted on HN: https://news.ycombinator.com/item?id=37176439
Besides, in this case, we would also have to upload 40 million photos to the cloud for Google to evaluate what’s on them.
YOLO is the best for such tasks; it works locally and is really fast.