OBSBot Records Five Years of 50% Growth
Shenzhen-based AI webcam maker OBSBot has grown 50% a year for five straight years and now holds more than 50% of the high-end webcam market, according to a KrASIA feature translated from a 36Kr profile of CEO Liu Bo. The company's Tiny series of AI-tracking webcams, launched in 2020, drove a recovery after a near-collapse in 2019 that included layoffs and inventory write-downs, per the report. OBSBot has since raised a "nine-figure USD" funding round from investors including Didi, HSG, Forebright Capital, CMB International, HKX, and Brizan Ventures, and now competes alongside established imaging players DJI and Insta360. Liu Bo told 36Kr the company survived its early years against dominant imaging incumbents largely through persistence rather than early profitability.
OBSBot's trajectory is a useful data point for edge-AI hardware teams: a narrow, differentiated computer-vision product, AI-powered subject tracking in a webcam, can capture a defensible niche against much larger incumbents, and the cash flow from that niche is what funds the harder, slower work of building an ecosystem. That pattern, more than the specific growth percentage, is the practitioner-relevant signal.
What happened
KrASIA, in a feature translated from a 36Kr profile of CEO Liu Bo, reports that OBSBot has grown 50% a year for five straight years and now holds more than 50% of the high-end webcam market, ranking first in that segment. The company's Tiny series, an AI-powered tracking webcam launched in 2020, drove a recovery after a 2019 downturn that included layoffs and inventory write-downs, according to the report. OBSBot has since raised a "nine-figure USD" funding round from investors including Didi, HSG, Forebright Capital, CMB International, HKX, and Brizan Ventures, and Liu Bo described surviving the company's near-revenue-less early years against established imaging giants such as DJI and Insta360.
Technical context
Commercializing a tracking webcam at this scale requires several engineering components relevant to practitioners: robust on-device person detection and pose estimation, smooth pan-tilt control loops, low-latency sensor fusion for subject re-acquisition, and video codecs optimized for constrained hardware. Sustaining leadership in the segment implies mature firmware-update tooling, continuous model calibration from field telemetry, and reproducible evaluation metrics for tracking quality, rather than reliance on large cloud-only model deployments.
Industry context
KrASIA frames OBSBot's growth as notable because it competes alongside established consumer-imaging players such as DJI and Insta360 while remaining a Shenzhen hardware startup. For ML and product teams, the takeaway is that edge-first camera products can still create durable commercial opportunities where software-defined differentiation, tracking, framing, developer APIs, is monetized alongside the hardware itself.
What to watch
Product line extensions and SDK releases that would signal developer-platform ambitions, further funding or partnership disclosures, and any public benchmarks or technical detail on the company's tracking algorithms beyond what this profile reports.
Key Points
- 1OBSBot has grown 50% annually for five years and holds over 50% of the high-end webcam market, per a verified KrASIA/36Kr profile of CEO Liu Bo.
- 2The company's on-device AI tracking, sensor fusion, and firmware pipelines illustrate the engineering demands of scaling edge computer-vision hardware.
- 3A nine-figure funding round from investors including Didi and HSG shows edge-AI hardware niches can still attract significant strategic capital.
Scoring Rationale
Re-verified this cycle: the specific 50%-growth, market-share, and nine-figure funding claims are now confirmed against the actual primary KrASIA/36Kr article (previously miscited to an unrelated 2024 piece). Score restored from 5.0 to 5.8 given the claims are properly sourced; still a solid rather than major story since it documents one company's growth narrative rather than a broader industry shift.
Sources
Primary source and supporting public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
