Remote zero-shot object recognition, i.e., offloading zero-shot object recognition tasks from one mobile device to remote mobile edge computing (MEC) server or another mobile device, has become a common and important task to conquer for 6G. With this goal, this paper first establishes a zero-shot multi-level feature extractor, which projects the image into the visual, semantic, as well as intermediate feature space in a lightweight way. Then, this paper proposes a novel multi-level feature transmission framework powered by a semantic knowledge base (SKB), and characterizes the semantic loss and required transmission latency at each level. Under this setup, this paper formulates the multi-level feature transmission optimization problem to minimize the semantic loss under the end-to-end latency constraint. Such a problem, however, is a multi-choice knapsack problem, and thus very difficult to solve. To resolve this issue, this paper proposes an efficient algorithm based on the convex concave procedure to find an efficient solution. Numerical results show that the proposed design outperforms the benchmarks, and illustrate the tradeoff between the transmission latency and zero-shot classification accuracy, as well as the effects of the SKBs at both the transmitter and receiver on classification accuracy.