From 10635ef94461b6c1b38508e65a04f8c7fea4e0f2 Mon Sep 17 00:00:00 2001 From: Vishnu Boddeti Date: Wed, 5 Aug 2026 19:29:31 -0400 Subject: [PATCH 1/9] Add face recognition benchmark pages --- README.md | 2 + face-recognition/Large.html | 163 ++++++++++++++++++++++++ face-recognition/Medium.html | 163 ++++++++++++++++++++++++ face-recognition/Single.html | 145 +++++++++++++++++++++ face-recognition/Small.html | 163 ++++++++++++++++++++++++ face-recognition/index.html | 239 +++++++++++++++++++++++++++++++++++ 6 files changed, 875 insertions(+) create mode 100644 face-recognition/Large.html create mode 100644 face-recognition/Medium.html create mode 100644 face-recognition/Single.html create mode 100644 face-recognition/Small.html create mode 100644 face-recognition/index.html diff --git a/README.md b/README.md index 85a6848..9652de6 100644 --- a/README.md +++ b/README.md @@ -32,6 +32,8 @@ Each workload represents a real-world use case with clear specifications, refere - [**BERT**](./BERT/index.html). Privacy-preserving BERT inference on encrypted inputs. Currently supports [BERT-Base (110M)](https://huggingface.co/google-bert/bert-base-cased-finetuned-mrpc) inference on the MRPC task in the [GLUE benchmark](https://gluebenchmark.com/). +- [**Face Recognition**](./face-recognition/index.html). Privacy-preserving face verification using encrypted facial embeddings, with single-pair and batched benchmark results. + *Additional workloads are under development. Check back for updates.* ## Suite Structure diff --git a/face-recognition/Large.html b/face-recognition/Large.html new file mode 100644 index 0000000..8583bc9 --- /dev/null +++ b/face-recognition/Large.html @@ -0,0 +1,163 @@ + + + + + + FHE Benchmarking Results - Face Recognition (Large Batch - 1,024) + + + + + + + + + +

FHE Benchmarking Results - Face Recognition (Large Batch - 1,024)

+

Timing and batched quality values are averages across 3 measurement run(s). Single-pair scores and labels are listed per run. Durations are wall time unless labeled as worker time.

+
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
SubmitterBandwidthQualityTiming (harness)Timing (server)
NameDateCPURAMOrionSlotsR/LRunsKeysModelInputResultFHE EERFHE TAR@1%FHE TAR@0.1%ArcFace EERArcFace TAR@1%ArcFace TAR@0.1%EER gapTAR@1% gapTAR@0.1% gapMax EER gapResultTotalOffline setupOnline evaluationDatasetKeygenModel prepInput genInput prepInput encComputeDecryptPostprocessArcFaceQualityTotal (wall)Setup (wall)Model I/ORuntime compileInput I/O (worker)Transport (worker)Inference (worker)Compute (wall)Backbones (worker)Normalize (worker)Inner product (worker)Mean / pair
CryptoFace2026-07-31----L328.8G-128.0G6.0G0.07230.84770.82030.06050.93550.93160.0117-0.0879-0.11130.1500PASS19.508h--295.9ms23.813s4.795s28.252s18.584m41.823m18.126h3.519m50.5ms18.064m3.3ms18.107h25.229m-----17.672h466.686h1.179h13.542m1.035m
+
+ + + diff --git a/face-recognition/Medium.html b/face-recognition/Medium.html new file mode 100644 index 0000000..c615e8c --- /dev/null +++ b/face-recognition/Medium.html @@ -0,0 +1,163 @@ + + + + + + FHE Benchmarking Results - Face Recognition (Medium Batch - 256) + + + + + + + + + +

FHE Benchmarking Results - Face Recognition (Medium Batch - 256)

+

Timing and batched quality values are averages across 3 measurement run(s). Single-pair scores and labels are listed per run. Durations are wall time unless labeled as worker time.

+
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
SubmitterBandwidthQualityTiming (harness)Timing (server)
NameDateCPURAMOrionSlotsR/LRunsKeysModelInputResultFHE EERFHE TAR@1%FHE TAR@0.1%ArcFace EERArcFace TAR@1%ArcFace TAR@0.1%EER gapTAR@1% gapTAR@0.1% gapMax EER gapResultTotalOffline setupOnline evaluationDatasetKeygenModel prepInput genInput prepInput encComputeDecryptPostprocessArcFaceQualityTotal (wall)Setup (wall)Model I/ORuntime compileInput I/O (worker)Transport (worker)Inference (worker)Compute (wall)Backbones (worker)Normalize (worker)Inner product (worker)Mean / pair
CryptoFace2026-07-31----L328.8G-32.0G1.5G0.08680.83950.82080.07210.92560.92090.0146-0.0861-0.10010.1500PASS5.434h--419.5ms31.390s4.486s7.575s5.199m11.051m5.046h59.294s31.4ms5.307m2.7ms5.027h25.181m-----4.594h121.440h17.884m3.438m1.077m
+
+ + + diff --git a/face-recognition/Single.html b/face-recognition/Single.html new file mode 100644 index 0000000..df4c7bd --- /dev/null +++ b/face-recognition/Single.html @@ -0,0 +1,145 @@ + + + + + + FHE Benchmarking Results - Face Recognition (Single - 1) + + + + + + + + + +

FHE Benchmarking Results - Face Recognition (Single - 1)

+

Timing and batched quality values are averages across 1 measurement run(s). Single-pair scores and labels are listed per run. Durations are wall time unless labeled as worker time.

+
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
SubmitterBandwidthQualityTiming (harness)Timing (server)
NameDateCPURAMOrionSlotsR/LRunsKeysModelInputResultScore(s)Label(s)TotalOffline setupOnline evaluationDatasetKeygenModel prepInput genInput prepInput encComputeDecryptPostprocessArcFaceQualityTotal (wall)Setup (wall)Model I/ORuntime compileInput I/O (worker)Transport (worker)Inference (worker)Compute (wall)Backbones (worker)Normalize (worker)Inner product (worker)Mean / pair
CryptoFace2026-07-31AMD EPYC 7502 32-Core Processor1007.7 GiB7e27281354f95L128.8G114.1G128.1M1.0M0.588810135.974m9.654m26.320m295.4ms7.608m2.041m380.5ms8.178s32.863s24.456m1.173m27.1ms--24.084m19.658m11.497m8.104m909.3ms382.7ms20.636m4.245m20.596m2.089s356.3ms4.245m
+
+ + + diff --git a/face-recognition/Small.html b/face-recognition/Small.html new file mode 100644 index 0000000..8f6249b --- /dev/null +++ b/face-recognition/Small.html @@ -0,0 +1,163 @@ + + + + + + FHE Benchmarking Results - Face Recognition (Small Batch - 128) + + + + + + + + + +

FHE Benchmarking Results - Face Recognition (Small Batch - 128)

+

Timing and batched quality values are averages across 3 measurement run(s). Single-pair scores and labels are listed per run. Durations are wall time unless labeled as worker time.

+
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
SubmitterBandwidthQualityTiming (harness)Timing (server)
NameDateCPURAMOrionSlotsR/LRunsKeysModelInputResultFHE EERFHE TAR@1%FHE TAR@0.1%ArcFace EERArcFace TAR@1%ArcFace TAR@0.1%EER gapTAR@1% gapTAR@0.1% gapMax EER gapResultTotalOffline setupOnline evaluationDatasetKeygenModel prepInput genInput prepInput encComputeDecryptPostprocessArcFaceQualityTotal (wall)Setup (wall)Model I/ORuntime compileInput I/O (worker)Transport (worker)Inference (worker)Compute (wall)Backbones (worker)Normalize (worker)Inner product (worker)Mean / pair
CryptoFace2026-07-31----L328.8G-16.0G768.1M0.07450.83990.83990.06310.93690.93690.0114-0.0970-0.09700.1500PASS2.892h--452.0ms27.339s4.569s3.879s2.692m5.686m2.688h39.273s26.3ms2.614m2.0ms2.671h25.127m-----2.238h57.547h9.270m1.778m1.049m
+
+ + + diff --git a/face-recognition/index.html b/face-recognition/index.html new file mode 100644 index 0000000..552ac23 --- /dev/null +++ b/face-recognition/index.html @@ -0,0 +1,239 @@ + + + + +FHE Benchmarking: Face Recognition Workload + + + + + +
+ Repository is found at + github.com/fhe-benchmarking/face-recognition +
+ +

FHE Benchmarking: Face Recognition Workload

+ +

Results - CelebA face verification

+ + +

Specification

+

The face recognition workload implements an encrypted face verification functionality. +The input is a pair of in-the-wild face images drawn from the CelebA dataset. +The goal is a binary decision: determine whether the two images depict the same person (a genuine pair) or two different people (an impostor pair). +For each pair the model computes, under homomorphic encryption, a similarity score between the two faces; only that score is decrypted. A genuine/impostor decision corresponds to thresholding this score, and the harness evaluates quality by sweeping thresholds to compute the Equal Error Rate and TAR@FAR (see below). +

+ +

The workload includes two interfaces for benchmark submitters to implement:

+ + + + + + + + + + + + + + + + + + + + + + + + +
SizePairs (N)
Small128
Medium256
Large1,024
+ +

Each variant samples N pairs from a fixed master set of 1,024 screened CelebA pairs (512 genuine, 512 impostor); the Large variant uses the full set. +Hence, there are a total of four variants of this workload: single inference and batch inference for each one of the three sizes. +Submitters need not implement all four, instead each submitter can implement and report the results of any subset.

+ +

Submission to the benchmarking suite must set the implementation parameters so as to achieve security level of at least 128 bits (against a semi-honest server). +Submitters must document their choice of parameters and explain why they believe that it meets the 128-bit security mandate. +The reference CryptoFace submission uses RNS-CKKS with ring dimension N = 216 and a sparse ternary secret (Hamming weight 192); its parameter selection and security analysis are described in [AB25].

+ +

Submissions are also required to meet a quality bar defined relative to a plaintext ArcFace [DGX+19] baseline that the harness evaluates on the same sampled pairs. +For batched inference, the encrypted model's Equal Error Rate (EER) must be no more than 0.15 above the ArcFace baseline EER on the same pairs. +EER is used because it is threshold-free and stable across batch sizes; the harness additionally reports the True Accept Rate at fixed False Accept Rates (TAR@FAR = 1% and 0.1%). +For single inference, the harness reports the encrypted similarity score for the pair. +

+ +

The face-recognition harness contains a script that can be called to run the implementation of submitters, that script accepts command-line arguments to specify which interface of what instance size to run. +The harness currently supports verification on CelebA pairs with a reference CryptoFace [AB25] submission (a patch-wise CNN with polynomial activations, evaluated under RNS-CKKS). In the future, this workload may be updated to support more models and more datasets.

+ +
+

+$ uv run python harness/run_submission.py -h
+usage: run_submission.py [-h] [--num_runs NUM_RUNS] [--seed SEED]
+                         [--clrtxt CLRTXT]
+                         {0,1,2,3}
+
+Run Face Verification FHE benchmark.
+
+positional arguments:
+  {0,1,2,3}            Instance size (0-single/1-small/2-medium/3-large)
+
+options:
+  -h, --help           show this help message and exit
+  --num_runs NUM_RUNS  Number of times to run stages 4-10 (default: 1)
+  --seed SEED          Random seed for reproducible pair sampling (default: 42).
+                       Fixed by default so all submissions sample identical pairs.
+  --clrtxt CLRTXT      Set to 1 to force rerun of the cleartext reference
+
+
+ +

Because the harness generates the evaluation pairs (submission-agnostic) before invoking any submission, and the seed defaults to a fixed value, every submission run with the default configuration is evaluated on the same pairs and is therefore directly comparable.

+ +You can find more details on the face-recognition Github repository. + + +
+

Bibliography

+
+

[AB25] +CryptoFace: End-to-End Encrypted Face Recognition. +Wei Ao and Vishnu Naresh Boddeti. +IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025. +

+
+
+

[DGX+19] +ArcFace: Additive Angular Margin Loss for Deep Face Recognition. +Jiankang Deng, Jia Guo, Niannan Xue, and Stefanos Zafeiriou. +IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019. +

+
+ + From 16ed0c6e8682d1de5a0b11fe9bea1ab948d77ffd Mon Sep 17 00:00:00 2001 From: Vishnu Boddeti Date: Thu, 6 Aug 2026 11:59:52 -0400 Subject: [PATCH 2/9] Refresh face recognition results --- face-recognition/Large.html | 76 +++++++++++++++++++----------------- face-recognition/Medium.html | 74 ++++++++++++++++++----------------- face-recognition/Single.html | 62 +++++++++++++++-------------- face-recognition/Small.html | 74 ++++++++++++++++++----------------- 4 files changed, 151 insertions(+), 135 deletions(-) diff --git a/face-recognition/Large.html b/face-recognition/Large.html index 8583bc9..8afc79b 100644 --- a/face-recognition/Large.html +++ b/face-recognition/Large.html @@ -34,11 +34,11 @@

FHE Benchmarking Results - Face Recognition (Large Batch - 1,024)

- + - + @@ -47,6 +47,7 @@

FHE Benchmarking Results - Face Recognition (Large Batch - 1,024)

+ @@ -79,6 +80,7 @@

FHE Benchmarking Results - Face Recognition (Large Batch - 1,024)

+ @@ -95,17 +97,18 @@

FHE Benchmarking Results - Face Recognition (Large Batch - 1,024)

- - - - - + + + + + + - - - + + + @@ -117,32 +120,33 @@

FHE Benchmarking Results - Face Recognition (Large Batch - 1,024)

- - - - - - - - - - - - - - - - - - - - - - - - - - + + + + + + + + + + + + + + + + + + + + + + + + + + +
SubmitterSubmitter Bandwidth Quality Timing (harness)Timing (server)Timing (server)
Name RAM Orion SlotsChunk R/L Runs Keys ArcFace Quality Total (wall)Process lifetime Setup (wall) Model I/O Runtime compile
CryptoFace2026-07-31----2026-08-06 11:57:25AMD EPYC 7502 32-Core Processor1007.7 GiB7e27281354f91050 L 3 28.8G-128.0G6.0G114.1G128.1G1.0G 0.0723 0.8477 0.8203 -0.1113 0.1500 PASS19.508h--295.9ms23.813s4.795s28.252s18.584m41.823m18.126h3.519m50.5ms18.064m3.3ms18.107h25.229m-----17.672h466.686h1.179h13.542m1.035m12.507h17.689m12.213h325.8ms8.320m9.364m19.437s22.180m46.576m10.688h3.428m47.4ms18.950m11.7ms10.644h10.677h22.475m13.842m8.478m51.956m9.143m663.424h10.270h662.592h42.366m7.560m36.105s
diff --git a/face-recognition/Medium.html b/face-recognition/Medium.html index c615e8c..497b84d 100644 --- a/face-recognition/Medium.html +++ b/face-recognition/Medium.html @@ -34,11 +34,11 @@

FHE Benchmarking Results - Face Recognition (Medium Batch - 256)

- + - + @@ -47,6 +47,7 @@

FHE Benchmarking Results - Face Recognition (Medium Batch - 256)

+ @@ -79,6 +80,7 @@

FHE Benchmarking Results - Face Recognition (Medium Batch - 256)

+ @@ -95,17 +97,18 @@

FHE Benchmarking Results - Face Recognition (Medium Batch - 256)

- - - - - + + + + + + - + - + @@ -117,32 +120,33 @@

FHE Benchmarking Results - Face Recognition (Medium Batch - 256)

- - - - - - - - - - - - - - - - - - - - - - - - - - + + + + + + + + + + + + + + + + + + + + + + + + + + +
SubmitterSubmitter Bandwidth Quality Timing (harness)Timing (server)Timing (server)
Name RAM Orion SlotsChunk R/L Runs Keys ArcFace Quality Total (wall)Process lifetime Setup (wall) Model I/O Runtime compile
CryptoFace2026-07-31----2026-08-05 19:34:36AMD EPYC 7502 32-Core Processor1007.7 GiB7e27281354f91050 L 3 28.8G-114.1G 32.0G1.5G258.6M 0.0868 0.8395 0.8208 -0.1001 0.1500 PASS5.434h--419.5ms31.390s4.486s7.575s5.199m11.051m5.046h59.294s31.4ms5.307m2.7ms5.027h25.181m-----4.594h121.440h17.884m3.438m1.077m3.418h21.038m3.067h415.5ms8.578m12.453m5.577s4.611m12.660m2.670h1.675m25.4ms4.779m64.8ms2.633h2.664h20.390m12.060m8.221m8.295m1.903m143.533h2.294h143.339h9.848m1.763m32.254s
diff --git a/face-recognition/Single.html b/face-recognition/Single.html index df4c7bd..de885d5 100644 --- a/face-recognition/Single.html +++ b/face-recognition/Single.html @@ -34,11 +34,11 @@

FHE Benchmarking Results - Face Recognition (Single - 1)

- + - + @@ -47,6 +47,7 @@

FHE Benchmarking Results - Face Recognition (Single - 1)

+ @@ -70,6 +71,7 @@

FHE Benchmarking Results - Face Recognition (Single - 1)

+ @@ -86,11 +88,12 @@

FHE Benchmarking Results - Face Recognition (Single - 1)

- + - + + @@ -99,32 +102,33 @@

FHE Benchmarking Results - Face Recognition (Single - 1)

- - - - - - - - - - - - + + + + + + + + + + + + - - - - - - - - - - - - - + + + + + + + + + + + + + +
SubmitterSubmitter Bandwidth Quality Timing (harness)Timing (server)Timing (server)
Name RAM Orion SlotsChunk R/L Runs Keys ArcFace Quality Total (wall)Process lifetime Setup (wall) Model I/O Runtime compile
CryptoFace2026-07-312026-08-05 19:34:36 AMD EPYC 7502 32-Core Processor 1007.7 GiB 7e27281354f951050 L 1 28.8G 1.0M 0.588810 135.974m9.654m26.320m295.4ms7.608m2.041m380.5ms8.178s32.863s24.456m1.173m27.1ms47.710m19.391m28.319m306.0ms8.837m10.548m588.6ms8.774s2.307m24.623m1.233m28.0ms --24.084m19.658m11.497m8.104m909.3ms382.7ms20.636m4.245m20.596m2.089s356.3ms4.245m3.6ms23.892m24.272m19.709m11.547m8.059m1.834s347.5ms19.623m4.183m19.583m2.077s376.2ms4.183m
diff --git a/face-recognition/Small.html b/face-recognition/Small.html index 8f6249b..ff9c6cc 100644 --- a/face-recognition/Small.html +++ b/face-recognition/Small.html @@ -34,11 +34,11 @@

FHE Benchmarking Results - Face Recognition (Small Batch - 128)

- + - + @@ -47,6 +47,7 @@

FHE Benchmarking Results - Face Recognition (Small Batch - 128)

+ @@ -79,6 +80,7 @@

FHE Benchmarking Results - Face Recognition (Small Batch - 128)

+ @@ -95,17 +97,18 @@

FHE Benchmarking Results - Face Recognition (Small Batch - 128)

- - - - - + + + + + + - + - + @@ -117,32 +120,33 @@

FHE Benchmarking Results - Face Recognition (Small Batch - 128)

- - - - - - - - - - - - - - - - - - - - - - - - - - + + + + + + + + + + + + + + + + + + + + + + + + + + +
SubmitterSubmitter Bandwidth Quality Timing (harness)Timing (server)Timing (server)
Name RAM Orion SlotsChunk R/L Runs Keys ArcFace Quality Total (wall)Process lifetime Setup (wall) Model I/O Runtime compile
CryptoFace2026-07-31----2026-08-05 19:34:36AMD EPYC 7502 32-Core Processor1007.7 GiB7e27281354f91050 L 3 28.8G-114.1G 16.0G768.1M129.3M 0.0745 0.8399 0.8399 -0.0970 0.1500 PASS2.892h--452.0ms27.339s4.569s3.879s2.692m5.686m2.688h39.273s26.3ms2.614m2.0ms2.671h25.127m-----2.238h57.547h9.270m1.778m1.049m2.240h19.243m1.919h291.4ms8.375m10.863m1.649s2.560m8.395m1.656h1.632m25.2ms3.149m4.6ms1.614h1.647h22.159m13.494m8.556m9.351m1.239m77.353h1.245h77.251h5.171m56.208s35.014s
From a62122bbf52501878201dbc5891e88ba6cc45285 Mon Sep 17 00:00:00 2001 From: Vishnu Boddeti Date: Thu, 6 Aug 2026 12:24:55 -0400 Subject: [PATCH 3/9] Add FHE learning resources --- README.md | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/README.md b/README.md index 9652de6..48a214c 100644 --- a/README.md +++ b/README.md @@ -86,6 +86,16 @@ For questions or to get involved: Visit the Homomorphic Encryption Standardization community: [homomorphicencryption.org](https://homomorphicencryption.org) +Additional tutorials and research: + +- [Computer Vision over Homomorphically Encrypted Data — CVPR 2025 Tutorial](https://fhe4cv.github.io/) +- [SecureRAG: End-to-End Secure Retrieval-Augmented Generation](https://hal.cse.msu.edu/papers/securerag-end-to-end-secure-rag/) +- [Homomorphically Encrypted Biometric Template Fusion and Matching](https://hal.cse.msu.edu/papers/homomorphically-encrypted-biometric-template-fusion-matching/) +- [AutoFHE: Automated Adaptation of CNNs for Efficient Evaluation over FHE](https://hal.cse.msu.edu/papers/autofhe-efficient-cnns-over-fhe/) +- [Secure Face Matching Using Fully Homomorphic Encryption](https://hal.cse.msu.edu/papers/secure-face-matching/) +- [HERS: Homomorphically Encrypted Representation Search](https://hal.cse.msu.edu/papers/hers-encrypted-image-search/) +- [HEFT: Homomorphically Encrypted Fusion of Biometric Templates](https://hal.cse.msu.edu/papers/heft-encrypted-biometric-fusion/) + ### Organizers Andreea Alexandru, Flavio Bergamaschi, Shruthi Gorantala, Shai Halevi @@ -100,4 +110,4 @@ Andreea Alexandru, Flavio Bergamaschi, Shruthi Gorantala, Shai Halevi --- -[Code of Conduct](./CODE_OF_CONDUCT.html) | © 2026 [HomomorphicEncryption.org](https://homomorphicencryption.org) | Last updated 2026-02-19 +[Code of Conduct](./CODE_OF_CONDUCT.html) | © 2026 [HomomorphicEncryption.org](https://homomorphicencryption.org) | Last updated 2026-08-06 From 494cb7514c3414925855de884da461154154244b Mon Sep 17 00:00:00 2001 From: Vishnu Boddeti Date: Thu, 6 Aug 2026 13:48:49 -0400 Subject: [PATCH 4/9] Add publication details to resources --- README.md | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/README.md b/README.md index 48a214c..b785c25 100644 --- a/README.md +++ b/README.md @@ -89,12 +89,12 @@ Visit the Homomorphic Encryption Standardization community: Additional tutorials and research: - [Computer Vision over Homomorphically Encrypted Data — CVPR 2025 Tutorial](https://fhe4cv.github.io/) -- [SecureRAG: End-to-End Secure Retrieval-Augmented Generation](https://hal.cse.msu.edu/papers/securerag-end-to-end-secure-rag/) -- [Homomorphically Encrypted Biometric Template Fusion and Matching](https://hal.cse.msu.edu/papers/homomorphically-encrypted-biometric-template-fusion-matching/) -- [AutoFHE: Automated Adaptation of CNNs for Efficient Evaluation over FHE](https://hal.cse.msu.edu/papers/autofhe-efficient-cnns-over-fhe/) -- [Secure Face Matching Using Fully Homomorphic Encryption](https://hal.cse.msu.edu/papers/secure-face-matching/) -- [HERS: Homomorphically Encrypted Representation Search](https://hal.cse.msu.edu/papers/hers-encrypted-image-search/) -- [HEFT: Homomorphically Encrypted Fusion of Biometric Templates](https://hal.cse.msu.edu/papers/heft-encrypted-biometric-fusion/) +- [SecureRAG: End-to-End Secure Retrieval-Augmented Generation](https://hal.cse.msu.edu/papers/securerag-end-to-end-secure-rag/) — *NeurIPS GenAI4Health Workshop, 2025*. [OpenReview](https://openreview.net/pdf?id=5uXACIHz6K) +- [Homomorphically Encrypted Biometric Template Fusion and Matching](https://hal.cse.msu.edu/papers/homomorphically-encrypted-biometric-template-fusion-matching/) — *IEEE Transactions on Biometrics, Behavior, and Identity Science, 2025*. [PDF](https://hal.cse.msu.edu/assets/pdfs/papers/2025-tbiom-homomorphically-encrypted-biometric-template-fusion-and-matching.pdf) +- [AutoFHE: Automated Adaptation of CNNs for Efficient Evaluation over FHE](https://hal.cse.msu.edu/papers/autofhe-efficient-cnns-over-fhe/) — *33rd USENIX Security Symposium, 2024*. [arXiv](https://arxiv.org/abs/2310.08012) · [ePrint](https://eprint.iacr.org/2023/162) +- [Secure Face Matching Using Fully Homomorphic Encryption](https://hal.cse.msu.edu/papers/secure-face-matching/) — *IEEE International Conference on Biometrics: Theory, Applications, and Systems, 2018*. [arXiv](https://arxiv.org/abs/1805.00577) +- [HERS: Homomorphically Encrypted Representation Search](https://hal.cse.msu.edu/papers/hers-encrypted-image-search/) — *IEEE Transactions on Biometrics, Behavior, and Identity Science, 2022*. [arXiv](https://arxiv.org/abs/2003.12197) +- [HEFT: Homomorphically Encrypted Fusion of Biometric Templates](https://hal.cse.msu.edu/papers/heft-encrypted-biometric-fusion/) — *IEEE International Joint Conference on Biometrics, 2022*. [arXiv](https://arxiv.org/abs/2208.07241) ### Organizers Andreea Alexandru, Flavio Bergamaschi, Shruthi Gorantala, Shai Halevi From 9fa40d25794d43b72c2daf144de52560d7445802 Mon Sep 17 00:00:00 2001 From: Vishnu Boddeti Date: Thu, 6 Aug 2026 13:56:09 -0400 Subject: [PATCH 5/9] Deduplicate resource links --- README.md | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/README.md b/README.md index b785c25..c0bcbee 100644 --- a/README.md +++ b/README.md @@ -89,12 +89,12 @@ Visit the Homomorphic Encryption Standardization community: Additional tutorials and research: - [Computer Vision over Homomorphically Encrypted Data — CVPR 2025 Tutorial](https://fhe4cv.github.io/) -- [SecureRAG: End-to-End Secure Retrieval-Augmented Generation](https://hal.cse.msu.edu/papers/securerag-end-to-end-secure-rag/) — *NeurIPS GenAI4Health Workshop, 2025*. [OpenReview](https://openreview.net/pdf?id=5uXACIHz6K) -- [Homomorphically Encrypted Biometric Template Fusion and Matching](https://hal.cse.msu.edu/papers/homomorphically-encrypted-biometric-template-fusion-matching/) — *IEEE Transactions on Biometrics, Behavior, and Identity Science, 2025*. [PDF](https://hal.cse.msu.edu/assets/pdfs/papers/2025-tbiom-homomorphically-encrypted-biometric-template-fusion-and-matching.pdf) -- [AutoFHE: Automated Adaptation of CNNs for Efficient Evaluation over FHE](https://hal.cse.msu.edu/papers/autofhe-efficient-cnns-over-fhe/) — *33rd USENIX Security Symposium, 2024*. [arXiv](https://arxiv.org/abs/2310.08012) · [ePrint](https://eprint.iacr.org/2023/162) -- [Secure Face Matching Using Fully Homomorphic Encryption](https://hal.cse.msu.edu/papers/secure-face-matching/) — *IEEE International Conference on Biometrics: Theory, Applications, and Systems, 2018*. [arXiv](https://arxiv.org/abs/1805.00577) -- [HERS: Homomorphically Encrypted Representation Search](https://hal.cse.msu.edu/papers/hers-encrypted-image-search/) — *IEEE Transactions on Biometrics, Behavior, and Identity Science, 2022*. [arXiv](https://arxiv.org/abs/2003.12197) -- [HEFT: Homomorphically Encrypted Fusion of Biometric Templates](https://hal.cse.msu.edu/papers/heft-encrypted-biometric-fusion/) — *IEEE International Joint Conference on Biometrics, 2022*. [arXiv](https://arxiv.org/abs/2208.07241) +- [SecureRAG: End-to-End Secure Retrieval-Augmented Generation](https://openreview.net/pdf?id=5uXACIHz6K) — *NeurIPS GenAI4Health Workshop, 2025*. +- [Homomorphically Encrypted Biometric Template Fusion and Matching](https://hal.cse.msu.edu/assets/pdfs/papers/2025-tbiom-homomorphically-encrypted-biometric-template-fusion-and-matching.pdf) — *IEEE Transactions on Biometrics, Behavior, and Identity Science, 2025*. +- [AutoFHE: Automated Adaptation of CNNs for Efficient Evaluation over FHE](https://arxiv.org/abs/2310.08012) — *33rd USENIX Security Symposium, 2024*. +- [Secure Face Matching Using Fully Homomorphic Encryption](https://arxiv.org/abs/1805.00577) — *IEEE International Conference on Biometrics: Theory, Applications, and Systems, 2018*. +- [HERS: Homomorphically Encrypted Representation Search](https://arxiv.org/abs/2003.12197) — *IEEE Transactions on Biometrics, Behavior, and Identity Science, 2022*. +- [HEFT: Homomorphically Encrypted Fusion of Biometric Templates](https://arxiv.org/abs/2208.07241) — *IEEE International Joint Conference on Biometrics, 2022*. ### Organizers Andreea Alexandru, Flavio Bergamaschi, Shruthi Gorantala, Shai Halevi From 784fd9e097748aa4ec42bf66bb8984e94c110133 Mon Sep 17 00:00:00 2001 From: Vishnu Boddeti Date: Thu, 6 Aug 2026 15:13:46 -0400 Subject: [PATCH 6/9] Order resources by publication year --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index c0bcbee..5ff9c3f 100644 --- a/README.md +++ b/README.md @@ -92,9 +92,9 @@ Additional tutorials and research: - [SecureRAG: End-to-End Secure Retrieval-Augmented Generation](https://openreview.net/pdf?id=5uXACIHz6K) — *NeurIPS GenAI4Health Workshop, 2025*. - [Homomorphically Encrypted Biometric Template Fusion and Matching](https://hal.cse.msu.edu/assets/pdfs/papers/2025-tbiom-homomorphically-encrypted-biometric-template-fusion-and-matching.pdf) — *IEEE Transactions on Biometrics, Behavior, and Identity Science, 2025*. - [AutoFHE: Automated Adaptation of CNNs for Efficient Evaluation over FHE](https://arxiv.org/abs/2310.08012) — *33rd USENIX Security Symposium, 2024*. -- [Secure Face Matching Using Fully Homomorphic Encryption](https://arxiv.org/abs/1805.00577) — *IEEE International Conference on Biometrics: Theory, Applications, and Systems, 2018*. - [HERS: Homomorphically Encrypted Representation Search](https://arxiv.org/abs/2003.12197) — *IEEE Transactions on Biometrics, Behavior, and Identity Science, 2022*. - [HEFT: Homomorphically Encrypted Fusion of Biometric Templates](https://arxiv.org/abs/2208.07241) — *IEEE International Joint Conference on Biometrics, 2022*. +- [Secure Face Matching Using Fully Homomorphic Encryption](https://arxiv.org/abs/1805.00577) — *IEEE International Conference on Biometrics: Theory, Applications, and Systems, 2018*. ### Organizers Andreea Alexandru, Flavio Bergamaschi, Shruthi Gorantala, Shai Halevi From 226fc1b6265102a51850e641422a63dcbdbfe257 Mon Sep 17 00:00:00 2001 From: Vishnu Boddeti Date: Thu, 6 Aug 2026 16:27:19 -0400 Subject: [PATCH 7/9] Fix local preview and result headers --- face-recognition/Large.html | 17 +++++++++-------- face-recognition/Medium.html | 17 +++++++++-------- face-recognition/Single.html | 17 +++++++++-------- face-recognition/Small.html | 17 +++++++++-------- index.html | 36 ++++++++++++++++++++++++++++++++++-- 5 files changed, 70 insertions(+), 34 deletions(-) diff --git a/face-recognition/Large.html b/face-recognition/Large.html index 8afc79b..d27f497 100644 --- a/face-recognition/Large.html +++ b/face-recognition/Large.html @@ -15,9 +15,10 @@ h1 { font-size: 1.5rem; letter-spacing: 0; } .table-wrap { width: 100%; overflow-x: auto; } #resultsTable { border-collapse: collapse; width: 100%; } - #resultsTable th, #resultsTable td { padding: 4px 8px; white-space: nowrap; } - #resultsTable thead tr:first-child th { background: #263645; color: white; text-align: center; } - #resultsTable thead tr:nth-child(2) th { color: white; font-size: 0.85rem; } + .table-wrap table.dataTable th, .table-wrap table.dataTable td { padding: 4px 8px; white-space: nowrap; } + .table-wrap table.dataTable thead tr:first-child th { color: white; text-align: center; } + .table-wrap table.dataTable thead tr:nth-child(2) th { color: white; font-size: 0.85rem; } + .group-header { text-align: center !important; vertical-align: middle; } .submitter-header { background: #2471a3 !important; } .bandwidth-header { background: #1e8449 !important; } .quality-header { background: #7d3c98 !important; } @@ -34,11 +35,11 @@

FHE Benchmarking Results - Face Recognition (Large Batch - 1,024)

- - - - - + + + + + diff --git a/face-recognition/Medium.html b/face-recognition/Medium.html index 497b84d..177e908 100644 --- a/face-recognition/Medium.html +++ b/face-recognition/Medium.html @@ -15,9 +15,10 @@ h1 { font-size: 1.5rem; letter-spacing: 0; } .table-wrap { width: 100%; overflow-x: auto; } #resultsTable { border-collapse: collapse; width: 100%; } - #resultsTable th, #resultsTable td { padding: 4px 8px; white-space: nowrap; } - #resultsTable thead tr:first-child th { background: #263645; color: white; text-align: center; } - #resultsTable thead tr:nth-child(2) th { color: white; font-size: 0.85rem; } + .table-wrap table.dataTable th, .table-wrap table.dataTable td { padding: 4px 8px; white-space: nowrap; } + .table-wrap table.dataTable thead tr:first-child th { color: white; text-align: center; } + .table-wrap table.dataTable thead tr:nth-child(2) th { color: white; font-size: 0.85rem; } + .group-header { text-align: center !important; vertical-align: middle; } .submitter-header { background: #2471a3 !important; } .bandwidth-header { background: #1e8449 !important; } .quality-header { background: #7d3c98 !important; } @@ -34,11 +35,11 @@

FHE Benchmarking Results - Face Recognition (Medium Batch - 256)

SubmitterBandwidthQualityTiming (harness)Timing (server)SubmitterBandwidthQualityTiming (harness)Timing (server)
Name
- - - - - + + + + + diff --git a/face-recognition/Single.html b/face-recognition/Single.html index de885d5..0bcc79e 100644 --- a/face-recognition/Single.html +++ b/face-recognition/Single.html @@ -15,9 +15,10 @@ h1 { font-size: 1.5rem; letter-spacing: 0; } .table-wrap { width: 100%; overflow-x: auto; } #resultsTable { border-collapse: collapse; width: 100%; } - #resultsTable th, #resultsTable td { padding: 4px 8px; white-space: nowrap; } - #resultsTable thead tr:first-child th { background: #263645; color: white; text-align: center; } - #resultsTable thead tr:nth-child(2) th { color: white; font-size: 0.85rem; } + .table-wrap table.dataTable th, .table-wrap table.dataTable td { padding: 4px 8px; white-space: nowrap; } + .table-wrap table.dataTable thead tr:first-child th { color: white; text-align: center; } + .table-wrap table.dataTable thead tr:nth-child(2) th { color: white; font-size: 0.85rem; } + .group-header { text-align: center !important; vertical-align: middle; } .submitter-header { background: #2471a3 !important; } .bandwidth-header { background: #1e8449 !important; } .quality-header { background: #7d3c98 !important; } @@ -34,11 +35,11 @@

FHE Benchmarking Results - Face Recognition (Single - 1)

SubmitterBandwidthQualityTiming (harness)Timing (server)SubmitterBandwidthQualityTiming (harness)Timing (server)
Name
- - - - - + + + + + diff --git a/face-recognition/Small.html b/face-recognition/Small.html index ff9c6cc..c5c096e 100644 --- a/face-recognition/Small.html +++ b/face-recognition/Small.html @@ -15,9 +15,10 @@ h1 { font-size: 1.5rem; letter-spacing: 0; } .table-wrap { width: 100%; overflow-x: auto; } #resultsTable { border-collapse: collapse; width: 100%; } - #resultsTable th, #resultsTable td { padding: 4px 8px; white-space: nowrap; } - #resultsTable thead tr:first-child th { background: #263645; color: white; text-align: center; } - #resultsTable thead tr:nth-child(2) th { color: white; font-size: 0.85rem; } + .table-wrap table.dataTable th, .table-wrap table.dataTable td { padding: 4px 8px; white-space: nowrap; } + .table-wrap table.dataTable thead tr:first-child th { color: white; text-align: center; } + .table-wrap table.dataTable thead tr:nth-child(2) th { color: white; font-size: 0.85rem; } + .group-header { text-align: center !important; vertical-align: middle; } .submitter-header { background: #2471a3 !important; } .bandwidth-header { background: #1e8449 !important; } .quality-header { background: #7d3c98 !important; } @@ -34,11 +35,11 @@

FHE Benchmarking Results - Face Recognition (Small Batch - 128)

SubmitterBandwidthQualityTiming (harness)Timing (server)SubmitterBandwidthQualityTiming (harness)Timing (server)
Name
- - - - - + + + + + diff --git a/index.html b/index.html index 63b33b7..00ec4b8 100644 --- a/index.html +++ b/index.html @@ -110,10 +110,40 @@ color: #666; font-size: 0.9em; } + + .local-preview-note { + padding: 12px 15px; + border-left: 4px solid #3498db; + background: #eef7fd; + } -
+
+

FHE Benchmarking Suite

+

Local website preview. Use the links below to open the newly added face-recognition pages.

+ +

Face Recognition

+

Privacy-preserving face verification using encrypted facial embeddings.

+ + +

Additional Tutorials and Research

+ +
SubmitterBandwidthQualityTiming (harness)Timing (server)SubmitterBandwidthQualityTiming (harness)Timing (server)
Name