This type of Knoevenhagel-Walter variation can be found in patent Patent GB1449540 (http://l2.espacenet.com/dips/viewer?PN=GB1449540&CY=gb&LG=en&DB=EPD)
E.g.,
3-Bromo-4-hydroxy-beta-nitrostyrene
3-Bromo-4-hydroxybenzaldehyde (20,1g) and n-propylamine (7,6g) were refluxed in dry benzene (150ml) using a Dean-Stark water separator for 3 hours. The benzene phase was concentrated to get the crude aldimine as an oily product. The aldimine (24,2g), nitromethane (24ml) and glacial acetic acid (72ml) were refluxed for 45 minutes. The reaction mixture was diluted with cold water. The solid product was filtered off and crystallised from aqueous ethanol to obtain the title product; m.p. 132-34 deg C.
It seems to be a good method for hydroxybenzaldehydes. Admittedly there is no report of the yields here but I can imagine they aren´t too shabby.
It is also used to prepare the BDB (J) nitrostyrene precursor, as it cannot be made directly from piperonal/nitropropane: Post 103792 (https://www.thevespiary.org/talk/index.php?topic=7493.msg10379200#msg10379200)
(Rhodium: "3,4-Methylenedioxyphenyl-2-nitro-1-butene", Chemistry Discourse)