We've recently replaced Hartley/Zisserman (which was getting a bit long in the tooth) with:

Photogrammetric Computer Vision: Statistics, Geometry, Orientation and Reconstruction by Wrobel and Förstner.

Thanks for the recommendation, ordered. Hartley is very valuable to know how we got to where we are, from hand tuned features and RANSAC to more modern full photometric optimization and ML. It answers a lot of "why don't we just... " type questions while still providing the foundations.

Wrobel and Förstner is definitely a different style, to some degrees the two books complement each other. But it focuses much more on the stats side and gaining an intuitive understanding, which to me is really important, and if I wanted RANSAC the algorithms in Hartley are out of date anyways.