After selling 5 million stabilizers, Hohem decided to develop a "detachable" camera.
In 2026, the handheld smart imaging track is at its most contradictory moment.
On one hand, market popularity keeps rising. According to IDC data, in the first quarter of 2026 alone, global shipments of handheld smart cameras reached 4.14 million units, a year-on-year increase of 33%, with sales exceeding 100 billion yuan; it is estimated that by 2030, the global market size will exceed 40 million units.
On the other hand, the competition pattern is rapidly solidifying. DJI ranks firmly first with 65% of shipments, and Insta360 ranks second with 22%, with the two accounting for nearly 90% of the total market share. In the general action camera comprehensive market, DJI and Insta360 account for about 43.8% and 41.2% of sales share respectively, with a combined total of 85%.
This is a track where leading brands have established absolute advantages.
But a subtle phenomenon has also emerged. The hotter the market and the more segmented the categories, the more hesitant ordinary users are in their choices. Users who buy a gimbal pocket camera for stable shooting and good image quality will find that the flexibility of viewing angles is limited; thumb cameras are light enough to be mounted on various positions, but their practicality in frequent daily scenarios is weak. Users have to make trade-offs and compromises between stability, portability, viewing angle and usage frequency.
How to eliminate such trade-offs has become the starting point of Hohem's product thinking. At IFA 2026, Hohem's first 2-in-1 pocket gimbal camera, Hohem Eyepic, made its official debut. It adopts a split structure, which can not only give full play to the stable shooting capability of handheld gimbal cameras, but also realize the flexible viewing angle of wearable cameras.
Founded 12 years ago, Hohem has successively launched a number of market hits such as the face-tracking smartphone stabilizer T1, the foldable smartphone stabilizer D1, and the separably controlled smartphone gimbal V3, with its global market share of action camera stabilizers reaching 70%. After selling stabilizers to more than 5 million users worldwide, Hohem has completed a strategic leap from an "imaging accessory manufacturer" to a "smart imaging brand".
The emergence of Hohem Eyepic also provides a new observation entry point. This is not a simple horizontal expansion of categories, but a systematic response to a core proposition. When hardware parameters gradually approach the physical upper limit, what kind of product form should an enterprise starting from gimbal technology adopt to meet the real life recording needs of ordinary users?
The more segmented the categories, the harder it is for users to choose
In the past few years, the category differentiation of handheld smart shooting hardware has almost reached its extreme.
Gimbal cameras focus on 3-axis anti-shake and stability to meet the shooting needs of shooting while walking. DJI's Pocket series is the iconic product of this category, and the cumulative sales of Pocket 3 alone have exceeded 10 million units. Action cameras firmly capture the group of extreme outdoor enthusiasts, with core advantages of drop resistance, waterproof performance and high durability. Insta360 X series can capture materials from a full 360° viewing angle, realizing the creative freedom of "shoot first, frame later". For wearable thumb cameras that compete for light weight in terms of gram weight, Insta360 GO series weighs only 35.5g, supports brim clamping and magnetic fixation, and can almost achieve non-inductive wearing.
Each category has its own strengths, and has its own sharp edge in its vertical scenario.
IDC's *Global Handheld Smart Camera Market Tracking Report* points out that handheld smart cameras are evolving from "niche equipment for extreme sports" to "the second imaging system for mass Vlog creation and daily recording", and the user group has expanded from niche extreme sports to the vast number of sports enthusiasts, Vlog creators and ordinary consumers who record daily life.
But the other side of differentiation is the natural fragmentation of user needs. As more and more segmented hardware emerges, the selection threshold and trial-and-error cost for ordinary consumers continue to rise.
Every type of device on the market has a clear capability bias: portability and image quality, anti-shake stability and camera position freedom, smooth main perspective recording and creative multi-angle shooting often cannot be achieved at the same time. Faced with a complex parameter system and professional technical terms, most ordinary users find it difficult to accurately judge the suitable model. It has become the norm to follow recommendations and change devices repeatedly, and eventually fall into the usage dilemma of "buying multiple devices and leaving most of them unused".
Behind this, there is actually a collision of two sets of contradictory product logic.
For the demand-oriented group represented by creators, their core demand is professional tools, which have strict requirements for high-definition image quality, extreme anti-shake and performance in special scenarios. They are willing to pay a premium for individual parameters, and will also be equipped with corresponding professional equipment according to different shooting scenarios. Segmented products perform their respective duties and can accurately solve vertical pain points. Therefore, the mainstream iteration logic of the industry is to continuously divide scenarios, refine categories, and maximize a single performance to the extreme.
The daily recording scenarios of ordinary users are completely different, such as casual shooting during travel, daily gatherings, and outdoor team building. Users' shooting needs switch in real time with the scenarios. Compared with extreme parameters, the public values the daily practicality of the equipment more: light and portable, ready to record at any time, no threshold to get started. What they need is not a professional device with full performance, but a complete "sufficient" and "easy to use" recording solution: no need to switch between multiple devices, no need to make trade-offs for scenarios, making daily recording calm and stress-free.
In other words, leading manufacturers are accelerating penetration into each other, competing for parameters, image quality, anti-shake and lightweight, but they have never made up for the all-round adaptation experience that the public needs most. This is an experience gap that cannot be filled by hardware stacking, and it is also the most critical and hidden incremental gap in the current handheld imaging track.
What Hohem has captured is precisely the possibility of jumping out of this set of one-dimensional competition logic.
"The opportunity we see is not to add another segmented category. On the contrary, a truly easy-to-use product can reduce the hesitation and trade-offs between different forms when users make purchases." This is the core judgment of the Hohem team on the track opportunity.
In a market fully educated by leading brands, the opportunity for new entrants is not to "make similar products stronger", but to dig out unfully responded user demands and "make a difference" in the cracks.
This kind of "difference" does not rely on subversive technological breakthroughs, but requires a return to re-understanding of user needs.
"2-in-1" is subtraction, "separable" is addition
Hohem did not choose to stack parameters in existing categories, but turned to a more simple proposition: from the moment the user picks up the device to the end of shooting and puts it down, which links in the entire recording process can be simplified.
This proposition falls into the specific product definition, pointing to a completely different direction from the mainstream path. While leading brands continue to increase investment in sensor size, lens specifications and computing power platforms, Hohem has chosen a counter-intuitive path: instead of doing addition, it does subtraction, using one device to cover two complete shooting forms.
At the hardware structure level, Hohem Eyepic proposes a quick-split structure of "magnet + buckle". The technical difficulty of this solution is to integrate the stable main perspective of handheld gimbal cameras and the free creative perspective of wearable thumb cameras into a single set of hardware, so that Eyepic has two complete sets of shooting capabilities at the same time, and can work independently in both combined and separated states.
In the combined state, the camera is tightly connected with the gimbal base through the buckle and magnetic structure, becoming a pocket gimbal camera with 3-axis mechanical anti-shake capability. The gimbal can real-time offset the picture shake caused by hand shaking, and output smooth and stable main perspective pictures in common scenarios such as handheld Vlog, travel recording, and parent-child follow shooting.
After the camera body is separated, it becomes a wearable thumb camera, which can be used as an independent shooting unit, supporting 10-meter waterproof and 10-meter wireless image transmission. With the brim clip for wearing, magnetically fixed on metal objects, or underwater shooting, it can easily achieve first-person perspective, low-angle shooting for pets, fixed-point camera positions and other shots that are difficult to capture with traditional handheld devices.
In addition, when the device is separated, the gimbal motor will automatically lock, and the 3-axis structure remains fixed. It is not only convenient for users to hold the camera body with one hand and complete real-time framing through the screen, but also prevents the motor from swinging unnecessarily and reduces power consumption in non-working states.
The combined state solves the problem of "stable shooting no matter how you shoot", and the separated state solves the problem of "where to place the camera to shoot". The two forms complement each other. Users do not need to make trade-offs between gimbal cameras and wearable cameras, or purchase multiple devices. They only need to switch the structural form of the same set of hardware to freely switch between the stable main perspective and creative free camera positions.
The core value of this design is that it transforms the original "pre-set shooting scenario before purchasing" into real-time "switch whenever you need". Users no longer need to predict shooting needs in advance and prepare multiple devices. They only need to flexibly switch the device form according to the current scenario to adapt to diversified daily recording needs.
At the level of product interaction experience, this subtraction thinking is reflected more thoroughly.
Eyepic is equipped with a 2.51-inch screen. The choice of this size has a practical consideration: the larger the screen, the more intuitive the framing and composition, and the easier it is to read the menu information. The icons on the screen are all labeled with text, and the menu structure is presented with visual logic.
In a large number of user surveys, Hohem's R&D team found that many novice users are prone to recognition barriers when facing a pure icon operation interface, and it is difficult to quickly understand the corresponding functions behind the icons. "Adding text annotations to icons does not have a high technical threshold, but few manufacturers are willing to do it." A relevant person in charge of Hohem told 36Kr that this type of optimization mainly serves non-professional users. Under the development logic oriented by hardware parameters, such experience demands are often easily ignored and placed in a secondary position in the product iteration process.
In terms of imaging specifications, Eyepic supports lossless 4K recording at 2160x3840@60fps in the vertical screen state, and it is also the first gimbal camera in the industry that currently supports 4K 60fps vertical screen recording. Based on Hohem's self-developed gimbal structure, the sensor rotates longitudinally when the screen rotates, which can sample the full sensor frame. Unlike secondary cropping of horizontal screen images that will lose a large number of pixels, the physical rotation solution retains all pixels of the sensor; at the same time, for social media users who are accustomed to vertical content distribution, the footage can be used directly after shooting, eliminating the step of post cropping and adaptation.
At the software level, the supporting APP of Eyepic has built-in editing templates customized for different scenarios. After users select them, the APP automatically completes material matching, rhythm beat alignment and transition generation. Users do not need to understand the operation logic of timeline editing, nor do they need to understand the parameter meanings of color and audio. They can export a finished video that can be shared directly with two steps, further reducing the full-link threshold from shooting to publishing.
From the hardware structure to the interactive interface and then to the software algorithm, these designs do not seem to show off skills, but they point to the same thing: converge complex calculation and structural logic to the inside of the device to the extreme, and leave simple operations to users.
In an industry dominated by "performance competition", investing resources and energy in "making products simpler" requires no less judgment than making a technological breakthrough.
This trade-off comes from Hohem's understanding of the underlying needs of different user groups. In the process of developing Eyepic, the team has been asking two core questions: What exactly are users paying for, and how should the priority of product definition be adjusted? In the era of excess parameters, how to make products that are closer to the real needs of mass users?
And Eyepic is exactly the answer Hohem gives: Let technology hide behind the product, and let recording return to instinct.
From gimbal to camera, Hohem's way to break through
After imaging becomes popular for all, the threshold of recording has dropped to a historical low. But a new problem has emerged: What kind of equipment do ordinary people need to complete daily recording?
In the past ten years, mobile phones have solved the problem of "carrying at any time", but failed to solve the problem of "stable shooting no matter how you shoot"; cameras have solved the problem of image quality, but have not solved the problem of "how to complete shooting alone"; various segmented cameras have met the needs of vertical scenarios, but virtually turned "selecting equipment" into the first threshold for ordinary users.
Hohem launched Hohem Eyepic at this node. It is not so much an extension from gimbals to cameras, but a company that has accumulated 12 years in the mobile shooting field, giving a systematic answer to the above questions.
The starting point of this answer began with an unfinished exploration seven years ago.
At the 2019 Hong Kong Consumer Electronics Show, Hohem once exhibited a prototype action camera integrating a 3-axis gimbal. This product did not eventually go into mass production. From an objective perspective, at that time, the team still had a lot to precipitate and improve in the R&D of imaging host and supporting supply chain. As a start-up company, limited resources must be prioritized to the core gimbal business.
However, this early exploration planted a long-term thinking direction within the team: Is it possible to use one device that not only has the stable main perspective of a gimbal camera, but also can reach shooting positions that ordinary handheld devices cannot cover?
This question was put on hold, but never disappeared.
In the following years, Hohem did not rush to make cameras. It chose a seemingly more circuitous but actually more solid path, repeatedly verifying the idea of "making the lens break away from the human hand" in the gimbal category. In 2021, it launched the AI smartphone gimbal V2 to realize local recognition and follow shooting; in 2024, the separably controlled smartphone gimbal V3 and the detachable monitoring gimbal M7 verified remote control shooting and real-time preview respectively.
These products have gained good reputation and sales in the market. More importantly, they have jointly completed one thing on the technical path: turning the logic of "separable lens and body, remote controllable, real-time preview" from a concept into mass-producible engineering capabilities.
