Here is a couple of examples from Plautus that could be relevant for your question:
a. Personal use of a transitive verb like decet:
contempla ut haec (vestis) me deceat (Pl. Most. 172). 'See how this dress suits me.'
b. Impersonal use of a transitive verb like decet:
ita ut vos decet (Pl. Most. 729). 'so as befits you'.
The following example could also be useful for you. Note that in this example fallit is a transitive verb that is also used impersonally:
nisi me fallit (Cic. Fam. 5, 20, 6), 'unless I am mistaken' ('unless it deceives me').
See A&G (1903: 388c: pp. 241-242) for other similar impersonal verbs that take accusative case.
From a more theoretical point of view, those impersonal verbs that take accusative case are expected to be grammatically marked (e.g., compared to the ones that take dative: cf. ita nobis decet (Ter. Ad. 928), 'thus it befits us'). Cf. the so-called "Burzio's Generalization", whereby, if a verb does not assign a semantic function to its subject, then it does not assign accusative case to its object.
Note also that Burzio's famous grammatical generalization appears to be infringed by the well-known constructions of the pudet me tui type since these verbs assign accusative case to the experiencer object but do not assign any semantic function to their subject (these verbs are impersonal). Interestingly, the accusative case of the experiencer argument of these stative psychological verbs has been originally related to those causative verbs of the 2nd conjugation. Cf. Matasović's (2013) proposal that "the irregular case-frames of the Latin bivalent statives are innovations based on the analogy with the case-frames of causative verbs, which had the Causee argument in the accusative case".
Note how this proposal somewhat restores Burzio's insight. The impersonal verbs at issue here assign accusative case to their experiencer object since they can be originally related to (not impersonal!) transitive verbs with causative meaning. As predicted by Burzio, those transitive verbs that assign accusative case are not (initially!) expected to be impersonal.
So, as for your question above ("Is it possible to form impersonal actives of transitive verbs?"), my preliminary answer is: yes, but only secondarily (e.g., through an analogical process. Cf. Matasović's (2013) proposal). NB: not primarily, because of Burzio's Generalization.