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

Spot the Drowning Child (2015)

spotthedrowningchild.com

Post reply on HN
Print view View original

81–90 of 424 posts

Re: Spot the Drowning Child (2015)

#81

Post by Fnoord »

ElgamalDSA wrote: post #25
HenryBemis wrote: post #11

To all parents out there, it takes 10 seconds for a toddler to drown. Once in the water, if they go under, they tend to get disoriented and don't know where is up or down. When you are with your kids near a swimming pool or the sea, keep your eyes glued on them. My rule is "never break line of sight" (I borrowed the term from Assassin's Creed where if you break line of sights from the guards chasing you for a few sec…

Friends of friends recently had their 1.5 year old drown in the bathtub when her dad left her alone for a minute. When you are at a pool / lake / sea with a kid that cannot swim, keep them in arms reach.

As a young adult, I almost drowned in a bathtub when taking magic mushrooms for the first time (a very small dosage of ~1/6th of a full dosage). I did put less water in the tub, but I still fell asleep, and when I woke up, the water line was a few mm from my nose openings. Do not use such drugs without a babysitter!

Re: Spot the Drowning Child (2015)

#82

Post by _ph_ »

JshWright wrote: post #40
chrismeller wrote: post #5

All in all after watching several of their videos I feel like I do a good job of recognizing the drowning person, but I’m amazed at how quickly the lifeguards spot it and dive into action. Even knowing that in this short clip there is absolutely someone drowning I still have doubt, but the lifeguard who doesn’t have that context is already half way to the person by the time I’m sure.

The guards actually have a lot more context. They have been watching the people in the pool and have mentally sorted them by how much attention they need. They have also been staring at the same pool for hours and their brain is ignoring all the visual noise that is distracting you. They also have the benefit of stereo vision and sound.

The sound of the videos is awful and little use, but I would guess, in reality, hearing can help too. And humans are exceptionally able to focus their attention on a single sound source.

In general, I would assume, like with driving, the job of the lifguard is a bit automatic, where the mind on its own filters out the ones, which might need stronger attention.

Re: Spot the Drowning Child (2015)

#83

Post by amelius »

tiborsaas wrote: post #80

What's the point of this? Should I be surprised that I do a poor job at watching a camera footage compared to a person being present there with all sensory inputs, context, trained to spot drowning people?

The point is that most people think that it is easy to spot drowning people.

Re: Spot the Drowning Child (2015)

#84

Post by wichert »

AllegedAlec wrote: post #45
philshem wrote: post #42

Rates of drowning in Europe[0][1] vary by more than an order of magnitude. I'd be curious about compulsory swimming lessons in schools, as is done here in Switzerland[2], and its correlation to the rate of drowning. [0] https://ec.europa.eu/eurostat/web/products-eurostat-news/-/D... [1] http://78.136.22.110/europe/info/switzerland/switzerland-dro... [2] https://en.wikipedia.org/wiki/Swimming_lessons#Switzerland

Netherlands reporting in: swimming lessons are spread over three competency group, and as far as I know, nearly all primary schools take time in the curriculum to do at least the first two competency levels before the pupils turn 7 or 8.

That might depend on the region. Here (Leiden) it is not common for schools to offer swimming lessons. The vast vast majority (I have no numbers, but I would expect it to be close to 100%) of children do take swimming lessons from age 4. Most of them until they reach the third (C) exam, which tends to take about 1.5 years.

Re: Spot the Drowning Child (2015)

#85

Post by Darkstryder »

b0rsuk wrote: post #62
minimonk wrote: post #36

One of the important lessons I learnt from a lifeguard is that movies depict a very inaccurate representation of drowning. The movies would have you believe that drowning is a violent and noisy event when in reality it is an inconspicuous and silent event. The victim cannot shout or call for help when they are struggling to keep their nose above the water level. Another important lesson I learnt that sometimes when s…

Do you know a single movie which accurately shows drowning?

Secondary drowning, while not shown on-screen, is an important plot point of the series The Affair.

Re: Spot the Drowning Child (2015)

#86

Post by rezic »

Jaruzel wrote: post #69
wycy wrote: post #20

It seems like spotting drowning children could be a good use case for computer vision, at least as a backup. The heuristics for a drowning child are pretty marked, but they're hard to spot for humans distracted by lots of other stimuli.

But who would you sue if the computer vision didn't spot your child drowning?

> as a backup

It'd be a tool that can warn the lifeguards in the event that they didn't notice the drowning, not the only thing that monitors the pool.

Re: Spot the Drowning Child (2015)

#87

Post by mcv »

AllegedAlec wrote: post #45
philshem wrote: post #42

Rates of drowning in Europe[0][1] vary by more than an order of magnitude. I'd be curious about compulsory swimming lessons in schools, as is done here in Switzerland[2], and its correlation to the rate of drowning. [0] https://ec.europa.eu/eurostat/web/products-eurostat-news/-/D... [1] http://78.136.22.110/europe/info/switzerland/switzerland-dro... [2] https://en.wikipedia.org/wiki/Swimming_lessons#Switzerland

Netherlands reporting in: swimming lessons are spread over three competency group, and as far as I know, nearly all primary schools take time in the curriculum to do at least the first two competency levels before the pupils turn 7 or 8.

A lot of primary schools don't do swimming lessons anymore. Fortunately nearly all Dutch kids get swimming lessons at a pool, generally before they'd get them at school anyway. Though that can cause problems for the one immigrant kid that doesn't have a diploma. There have been at least two drownings during school swimming lessons in recent years.

Re: Spot the Drowning Child (2015)

#88

Post by zentiggr »

Jaruzel wrote: post #69
wycy wrote: post #20

It seems like spotting drowning children could be a good use case for computer vision, at least as a backup. The heuristics for a drowning child are pretty marked, but they're hard to spot for humans distracted by lots of other stimuli.

But who would you sue if the computer vision didn't spot your child drowning?

I think the more likely scenario is who wouldn't you sue?

Even the camera manufacturer wouldn't be immune from defending themselves.

Welcome to America.

Re: Spot the Drowning Child (2015)

#89

Post by _ph_ »

I am impressed by these videos any time the pop up on hacker news. But one thing struck me: that they are using those large floatation rings. A lot of the incidents seem to be where a child looses contact to the ring and then cannot swim on itself. I am wondering, why they are allowed at all. In my personal experience, I have rarely seen such rings in public pools and that basically means, you are not getting far into the deep part of the pool without some basic swimming skills. Most people/children wouldn't even try as they don't feel comfortable with deep water without an aid.

Re: Spot the Drowning Child (2015)

#90

Post by mimimi31 »

Why are there so many people in the deep end of the pool who don't know how to swim properly? Looking at some of the videos, it seems like the majority can't do a breaststroke and drowns if they can't dog paddle to anything buoyant within a few seconds.
Post reply on HN