First- and second-instance courts

Supreme Court

In a recent design patent infringement case involving electric bicycle and motorcycle designs, the opinions of the first-instance court,(1) the second-instance court(2) and the Supreme Court(3) showed that the court's opinion on which features are considered to be "salient" has a decisive impact on the result of the judgment.


According to article 136 of the Patent Act, the patentee of a design patent has the exclusive right to prevent others from exploiting its design or using similar designs without its consent – doing so would constitute design patent infringement. However, it can be difficult to determine whether two designs are "similar" in terms of their overall impression when it comes to an article such as a vehicle, the overall appearance of which is made up of many parts.

Most cases follow the Taiwan Intellectual Property Office's (TIPO's) 2016 guidance document "Essentials for Patent Infringement Determination", which stipulates that the determination of whether the overall visual impressions of two designs may confuse people should be made:

  • from the perspective of an ordinary consumer;
  • by observing the overall designs and evaluating them comprehensively; and
  • based on salient features – that is, the "parts or features that can easily attract an ordinary consumer's attention", including:
    • the "design features of the patent at dispute that are obviously different from prior art"; and
    • the "parts that are readily visible in normal use".

However, the selection and weighting of salient features varies from person to person.

First- and second-instance courts

The first- and second-instance courts considered that the appearance of the following features, which comprised the main body of the bike, was identical or similar in the design patent and the product at issue:

  • the front end;
  • the pedals (and the X-shaped pattern thereon);
  • the seat cushion;
  • the storage compartment; and
  • the seat post.

However, the first-instance court held that the design patent at issue and the accused infringing product were neither identical nor similar to each other in terms of their overall appearance. The court pointed out that the design patent at issue featured an inverted U-shaped rear fork rack, which gave the impression that the seat cushion and the storage box were suspended. This was obviously not identical to the shape of the shock absorber and the rear fork rack of the product at issue. In addition, the design at issue and the accused product had different handlebar, headlight and basket designs. These features were essential and frequently used, and thus an ordinary consumer would notice the differences between them easily.

The second-instance court, on the other hand, held that there was only a slight difference between the tilt angle of the shock absorber of the accused product and the upper end of the inverted U-shaped rear fork rack of the patent at issue. Also, the shock absorber of the accused product was installed to perform shock absorption – similar features had long been seen in prior art. The second-instance court further held that the design of the frame was a salient feature of the motorcycle model and thus was the core of the design.

In addition, the court held that the common features of the accused product and the patent at issue resided in the main frame and the seat cushion, which took up a large visual area and were readily visible in normal use. Thus, according to the court, they were sufficient to affect the overall visual impression of an ordinary consumer. The different features were mainly situated at the front and rear end portions of the bikes, which were further apart and took up a smaller portion of the overall visual area. Thus, they did not affect the overall visual impression. As a result, the second tribunal held that patent infringement was established.

In determining patent infringement, the second-instance court went a step further than the first-instance court by performing a three-way visual comparison test (Figure 1). The court compared the accused product with various prior art references to understand the extent of similarity between the patent at issue, the accused product and the prior art references and thereby determined whether the accused product and the patent at issue were similar in appearance. Ultimately, the second-instance court concluded that the appearance of the accused product was closest to that of the patent at issue. This comparison test enabled the court to quickly and visually determine whether the accused infringing product had adopted the features of the patent at issue that were different from the prior art.

Figure 1: example of the second-instance court's three-way visual comparison test

Supreme Court

In a manner close to a determination and judgment of the facts, the Supreme Court reversed the second-instance court's decision. In particular, the Supreme Court noted that one of the major differences between the patent at issue and the reference cited in the patent at issue (ie, the prior art reference cited by the TIPO when the patent at issue was under TIPO's examination prior to being granted) (D100737) lay in the special visual effect formed by the U-shaped rear fork rack, together with other features. Like the first-instance court, the Supreme Court indicated that the front and rear ends and the headlights (ie, not just the frame) of the bike would be readily visible to an ordinary consumer at the time of purchase and use.


The overall difference in appearance (ie, not just the appearance of the rear fork rack) between the prior art D100737 design patent (on the left in Figure 2) cited by the Supreme Court and the patent at issue (on the right in Figure 2) is significant. Although the U-shaped fork rack of the patent at issue was indeed a novel feature compared with D100737, another prior art reference (D104904) (on the left in Figure 3) had already disclosed a U-shaped fork rack design according to the diagram attached to the interlocutory judgment of the second-instance court. Therefore, it is questionable that the U-shaped fork rack seen in the D104904 prior art reference should be given a greater weight in the infringement comparison.

Figure 2: a comparison of the D100737 design patent and the design patent at issue

Figure 3: a comparison of the D104904 design patent and the design patent at issue

The decisions in this case show that when determining whether design patents – particularly those with multiple components – have been infringed, the courts may arrive at completely different conclusions, depending on the points on which they focus. Even though finding common and distinguishing features is an approach that the courts must take in such judgments, the determination of whether a design patent has been infringed cannot depart from the principle that the comparison must be "overall".

To avoid the inconsistency of judgments caused by the overly subjective selection of which features are salient, is it possible to establish an objective standard for determining whether a design patent has been infringed in accordance with the fundamental theory of design and under the principle of "overall comparison"? Significant work remains to be done by parties involved in litigation and the courts in order to achieve this goal.

For further information on this topic please contact May Chen or Andy Hsieh at Lee and Li Attorneys at Law by telephone (+886 2 2763 8000) or email ([email protected] or [email protected]). The Lee and Li Attorneys at Law website can be accessed at


(1) 2019 Min Zhuan Su Zi No. 31.

(2) 2020 Min Zhuan Shang Zi No. 6.

(3) 2022 Tai Shang Zi No. 487.