Also pressure proofing cameras are not cheap I presume.
You can see it in the last Smarter Every Day video[1] where they're talking about surfacing a submarine in the arctic ocean.
At several points you can see a video monitor showing an upward view of the ice.
[1] https://youtu.be/XFJnWp1tAdU?t=1427
e: to be clear, that upward looking camera isn't going to be enough to detect much beyond your own footprint. If someone is about to cross that area you'd need some other way of detecting them.
This was a contributing factor to a collision in the Mediterranean about a decade or so ago.
From what I overheard in cafeterias frequented navy navy men while living in Vladivostok is that active sonar is used all the time unless deep blue waters.
Navigational active sonars are nowhere near as detectable as long range military ones.
It's the "flashlight in the dark room" problem. Your sonar ping has to travel to an object and reflect back to the submarine. But an adversary just needs to hear the original ping to know what direction it came from, so you can be detected at a longer distance, and potentially a much longer distance, than you could reliably detect the other submarine at.
That's not to say active sonar is never used, but at least with the U.S. it's not going to be continuously used, and it's going to be rare to use it if you're in a mission area.
A sonar transducer that can emit a ping with directionality is sometimes useful (e.g. a fathometer to measure depth to the ocean bottom, since the crushing pressure of the depths ensures the area below you is generally clear of enemy subs), and those are used. But even there, that wouldn't be useful for navigating when otherwise blind; you use fathometers to confirm that you're approximately in the area you think you're in based on the chart readings.
I mean the difference is possibly hundreds of metres to hundreds of kilometres in between navigational sonars, vs sub detection sonars.
