Industries / Consumer goods / Eyewear
Frames
Designing and making the part of a pair of glasses that holds the lenses and sits on the face: the front, the bridge, the hinges and the side arms, called temples.
7
Milestones achieved, from 1280s to 2012.
2
To watch for: announced, not yet confirmed outside.
3
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1280s
Lenses riveted at the handles
The earliest spectacles were two mounted lenses riveted together at the ends of their handles. Their maker is unknown. A chronicle of the convent of Santa Caterina in Pisa says Friar Alessandro della Spina made eyeglasses that "another had first devised" and taught everyone how to build them, while the first maker kept the method secret.
LessonAn inventor can be made up to flatter a city. From 1684 Florentine historians credited Salvino degli Armati, quoting an epitaph that Baldanzi and Farini show to be false, and the story is still repeated.
Credit
an unknown maker in northern Italy; Alessandro della Spina as an early maker who shared the method
1730
Arms that grip the head
Rigid side arms of iron or steel, called temples, gripped the sides of the head and held spectacles on without a hand. The College of Optometrists says the London optician Edward Scarlett "is credited" with this modern style, and the Cowper and Newton Museum dates it to around 1730. Steel spectacles with double-pivoted side pieces, made between 1750 and 1800, survive at the Whipple Museum in Cambridge.
college-optometrists.orgcowperandnewtonmuseum.org.ukcollection.sciencemuseumgroup.org.ukcollections.whipplemuseum.cam.ac.uk
Credit
Edward Scarlett, whose trade card names him optician to King George II
1878
A frame factory in Cadore
Frame making in Cadore, in the Italian province of Belluno, moved from workshops into a factory when Angelo Frescura, with Giovanni Lozza and Leone Frescura, set one up at Calalzo di Cadore. The text of "The Lens of Time", an exhibition backed by the Italian eyewear makers' association ANFAO and the Eyewear Museum of Pieve di Cadore, calls it the first Italian eyewear factory. Luxottica began in Agordo in 1961, when Leonardo Del Vecchio set up a company making parts for the optical industry.
Credit
Angelo Frescura, Giovanni Lozza and Leone Frescura
1905
Sabae turns to frames
Masunaga Gozaemon brought spectacle craftsmen from Osaka to Asouzu village in Fukui Prefecture, Japan, and frame making began there in 1905. The prefecture's industry body, the Fukui Optical Association, says the world's first lightweight titanium frames were made there in 1981. The news site nippon.com says Sabae craftsmen were the first in the world to develop the technique in the 1980s, that titanium frames are much less likely to cause metal allergies, and that Sabae makes 96 percent of Japan's frames.
LessonBeing first on technique did not protect volume. nippon.com reports that Sabae's industry declined after 1992 as low-cost Chinese makers grew.
Credit
Masunaga Gozaemon; Sabae's frame makers
1930s
Plastic that could burn
The first plastic frames, sold from the 1910s and common by the 1930s, were made of cellulose nitrate. It softens at 65 °C and ignites at 70 °C, close together for a material that opticians heat to adjust; the UK opticians' body ABDO tells of a trainee left holding the smouldering remains of a frame. Cellulose acetate replaced it and is now the usual material for plastic frames. Nitrate is banned in the UK but still used in some developing countries, ABDO reports.
LessonThe fix is now written into the frame standard. ISO 12870 requires that a frame stops burning once the test rod is taken away.
Credit
the frame trade as a whole; no single maker is named in the sources found
1997
Frames get a standard
ISO 12870, first published on 15 December 1997, set durability tests for frames and tolerances for the sizes marked on them. Sizes are measured with the boxing system of ISO 8624 (1991), which draws a rectangle around each lens shape. The tolerances are ±0.5 mm for lens size and bridge width and ±2.0 mm for temple length, and a frame may not bend out of shape by more than 5 mm after 500 flexing cycles. The 2025 edition, published by BSI on 31 August 2025, also covers frames made by 3D printing. Metal frames are tested for nickel release, a common cause of skin allergy, with EN 16128, a method first issued by the European standards body CEN in 2011 and revised in 2015, against the EU limit of 0.5 µg/cm² a week. This came after EU ministers struck spectacle frames out of the draft nickel directive in December 1993, UK Hansard records.
cdn.standards.iteh.aiknowledge.bsigroup.comintertek.comapi.parliament.uk
Credit
ISO's committee on ophthalmic optics (ISO/TC 172/SC 7); CEN
2012
Frames printed from powder
MYKITA's Mylon frames are built by selective laser sintering, in which a laser fuses polyamide (nylon) powder layer by layer, as the trade site 3D Printing Industry reported in November 2012. In September 2016 HOYA Vision Care, with Materialise and the Hoet design studio, launched Yuniku, frames tailored to a scan of the wearer's face, which HOYA called the world's first 3D tailored eyewear. Materialise describes a 10,000 m² laser-sintering centre in Wrocław, Poland, with 30 machines, and still finishes its printed frames by hand.
Credit
MYKITA; HOYA Vision Care, Materialise and Hoet
Watch for
Acetate scrap made new
Re|Ace says it turns acetate scrap into 100 percent recycled acetate, and that the process can be repeated up to seven times. The claim matters because making frames from sheet acetate scraps about 80 percent of the material, by figures reported by FashionUnited in 2020 and by VoxelMatters in 2026.
Achieved whenan independent body certifies the recycled content and the repeat-cycle claim, and frames made from the material by another brand pass ISO 12870 and go on sale.
Claimed by
Re|Ace, as reported by VoxelMatters at the MIDO trade fair, February 2026
Frames from plant-based nylon
Arkema says its Rilsan Clear Rnew G850 and G820 resins for moulded frames take up to 62 percent of their carbon from plants, and that its Rilsan PA11 Fine powders for 3D printing are 100 percent bio-based. It says Acuitis won its children's frame design challenge, run with Materialise, and that the winning design will become a collection.
Achieved whenthe Acuitis collection is on sale and an outside lab confirms the bio-based carbon share, for example by radiocarbon testing to ASTM D6866.
Claimed by
Arkema, January 2026
Open
01
Robot polishing of acetate fronts
A robot cell that buffs CNC-cut acetate frame fronts to a gloss finish, in place of days of tumbling and hand buffing, and a published comparison of 30 robot-finished fronts with 30 finished by hand.
The briefCredit · unclaimed02
Checking the temple size
An open dataset of the sizes marked on frames (lens size, bridge width and temple length) against the sizes measured, for at least 200 frames from at least 20 brands, scored against the ISO 12870 tolerances.
The briefCredit · unclaimed03
Telling frame plastics apart
A handheld near-infrared (NIR) scanner and classifier that tells acetate, propionate, nitrate, nylon blends such as TR90 and other frame plastics apart in seconds, so used frames can be sorted for material recycling.
The briefCredit · unclaimed
Open 01 · the brief
Robot polishing of acetate fronts
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A robot cell that buffs CNC-cut acetate frame fronts to a gloss finish, in place of days of tumbling and hand buffing, and a published comparison of 30 robot-finished fronts with 30 finished by hand.
Why it is still open
Finishing acetate is still slow and partly manual. At Lowercase, a New York frame maker, a UR5 robot arm tends the milling machine, but each front then goes through four barrel-tumbling stages of 22 hours each and a manual buffing step (Cutting Tool Engineering, 2017). ABDO describes up to four tumbling stages of about 24 hours each, and Materialise lists hand finishing for its printed frames. A search found a 2019 paper on robot polishing tested on one eyeglass frame, which could not be opened, and no report of robot finishing of acetate fronts.
ctemag.comabdo.org.ukkramerindustriesonline.commaterialise.com
Built with what exists
A collaborative robot arm such as a Universal Robots UR5e, or low-cost SO-101 arms run with Hugging Face's open LeRobot software (see the Note "No download"); a control model trained from teleoperated demonstrations, such as NVIDIA's GR00T N1.7, which LeRobot supports, or GR00T 2 once NVIDIA releases its code (see the Note "Five demonstrations per task"); a variable-speed buffing wheel; a camera; a handheld gloss meter and a surface roughness gauge.
First step
Ask a small acetate workshop for 20 unpolished fronts, measure gloss and roughness after each stage of its normal process, and record 50 teleoperated buffing passes.
Done looks like
30 robot-finished and 30 hand-finished fronts of the same model, with gloss and roughness published for each; the robot's fronts within a margin set in advance of the hand-finished ones; every front still passing the ISO 12870 tests; and the finishing time per batch reported.
Open 02 · the brief
Checking the temple size
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
An open dataset of the sizes marked on frames (lens size, bridge width and temple length) against the sizes measured, for at least 200 frames from at least 20 brands, scored against the ISO 12870 tolerances.
Why it is still open
No published audit of marked against measured sizes was found. Testing labs such as QIMA measure frames with tracers and calipers to ISO 8624 for paying clients and do not publish the results. ISO 12870 allows ±0.5 mm on lens size and bridge width and ±2.0 mm on temple length (1997 edition).
Built with what exists
Digital calipers reading to 0.01 mm; a flatbed scanner; the open OpenCV library to fit the boxing rectangle to each scanned lens shape; a spreadsheet published under an open license.
First step
Write the measuring method and measure 20 frames twice, a week apart.
Done looks like
200 frames from 20 brands, each measured twice with repeat readings within 0.1 mm of each other, and the share outside tolerance published by brand and by where the frame was bought.
Open 03 · the brief
Telling frame plastics apart
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A handheld near-infrared (NIR) scanner and classifier that tells acetate, propionate, nitrate, nylon blends such as TR90 and other frame plastics apart in seconds, so used frames can be sorted for material recycling.
Why it is still open
Used frames are sorted by hand into broad groups. TerraCycle sorts optical waste into rigid plastics, metals, flexible plastics and glass, and Recycline separates metal from plastic, which ends up as plastic lumber (ABDO, 2022). ABDO's guide to frame materials says nitrate looks very similar to acetate and that propionate is hard to tell apart, and its identification tests rely on smell and on how the joints are pinned. Recycled acetate has buyers: Eastman's Acetate Renew and EOE Eyewear's REGRIND both use it. No public tool or study for sorting frames by infrared was found; Thermo Fisher describes infrared for quality control of frame plastics, not for sorting.
Built with what exists
A compact NIR spectrometer such as the Avantes AvaSpec-Mini-NIR, which Avantes sells for recycling work, or a trinamiX mobile NIR unit; reference labels confirmed by FTIR, the infrared method university chemistry labs use; a small classifier written in Python with scikit-learn.
First step
Collect 50 frames of known material from a frame maker's offcuts and a recycling bin, and record their reference spectra. Use no heat or burn tests, because nitrate is flammable.
Done looks like
At least 95 percent correct on 300 held-out frames whose material FTIR has confirmed, in under 5 seconds a frame, with spectra and code published.