Offline product recommendation method, device and equipment and readable storage medium

The invention relates to the field of artificial intelligence, and discloses an offline product recommendation method, which comprises the following steps: performing video recording on behavior actions of a user to obtain a user activity video; analyzing the user activity video to obtain the number of times and the watching duration of each product in a product counter watched by the user; analyzing the facial expression category of each product in the watching counter of the user; according to the number of times of watching each product in the counter by the user, the watching duration and the facial expression category of each product in the counter by the user, analyzing the interesting probability of each product in the counter by the user; and according to the interested probability, informing sales personnel to carry out product recommendation on the user. The invention also relates to a block chain technology, and the user activity video can be stored in a block chain node. The invention further prov