SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2407.13974 · ECCV · 2024

Continual Learning for Remote Physiological Measurement: Minimize Forgetting and Simplify Inference

Yan Chen, Yang Hu, Qian Liang

arXiv · PDF · Open in the Atlas

Code that ran

7 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 4 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 3 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2407.13974.json · badge: /api/ran/2407.13974.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
MayYoY/rPPGDIL canonical (classifier label)
classifier basis: census
7 ran · 1 no recorded run
FunctionStatusWhere it lives
cal_metric Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
MayYoY/rPPGDIL/evaluate/metrics.py
code served (permissive licence) · get_code("aa83f259724c908b")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create simple numpy arrays for predictions and labels
# Using distinct shapes to make dimensions clear
pred_phys = np.array([1.2, 2.1, 3.5, 4.8, 5.2, 6.1])
label_phys = np.array([1.0, 2.0, 3.0, 5.0, 5.0, 6.0])

# Define methods to use (or let it default to all)
methods = ['Mean', 'Std', 'MAE', 'RMSE', 'MAPE', 'R']

ARGS = {
    'pred_phys': pred_phys,
    'label_phys': label_phys,
    'methods': methods
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
detrend Ran
ran, honoured the contract we drafted
MayYoY/rPPGDIL/evaluate/postprocess.py
code served (permissive licence) · get_code("74b59eab5b0e9530")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family float, shape (2, 5) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create a small signal for detrending
# signal should have shape (..., T) where T is the time dimension
# The function will operate on the last dimension
signal = np.random.randn(2, 5)  # batch_size=2, T=5

ARGS = {
    "signal": signal,
    "lambda_value": 100
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (2, 5) — shape (2, 5) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
fft_physiology Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
MayYoY/rPPGDIL/evaluate/postprocess.py
code served (permissive licence) · get_code("7d7d0afab3889569")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 1, dtype family float, shape (2,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
import scipy.signal

def next_power_of_2(x):
    return 1 if x == 0 else 2 ** (x - 1).bit_length()

def detrend(signal, lambda_value=100):
    from scipy.sparse import spdiags
    T = signal.shape[-1]
    H = np.identity(T)
    ones = np.ones(T)
    minus_twos = -2 * np.ones(T)
    diags_data = np.array([ones, minus_twos, ones])
    diags_index = np.array([0, 1, 2])
    D = spdiags(diags_data, diags_index, T - 2, T).toarray()
    designal = (H - np.linalg.inv(H + lambda_value ** 2 * D.T.dot(D))).dot(signal.T).T
    return designal

def fft_physiology(signal: np.ndarray, target='pulse', Fs=30, diff=True, detrend_flag=True):
    if diff:
        signal = signal.cumsum(axis=-1)
    if detrend_flag:
        signal = detrend(signal, 100)
    if target == 'pulse':
        [b, a] = scipy.signal.butter(1, [0.75 / Fs * 2, 2.5 / Fs * 2], btype='bandpass')
    else:
        [b, a] = scipy.signal.butter(1, [0.08 / Fs * 2, 0.5 / Fs * 2], btype='bandpass')
    signal = scipy.signal.filtfilt(b, a, np.double(signal))
    N = next_power_of_2(signal.shape[-1])
    freq, psd = scipy.signal.periodogram(signal, fs=Fs, nfft=N, detrend=False)
    if target == 'pulse':
        mask = np.argwhere((freq >= 0.75) & (freq <= 2.5))
    else:
        mask = np.argwhere((freq >= 0.08) & (freq <= 0.5))
    freq = freq[mask]
    if len(signal.shape) == 1:
        idx = psd[mask.reshape(-1)].argmax(-1)
    else:
        idx = psd[:, mask.reshape(-1)].argmax(-1)
    phys = freq[idx] * 60
    return phys.reshape(-1)

# Create small test signal: batch of 2 signals, each with 16 samples
np.random.seed(0)
signal_data = np.random.randn(2, 16).astype(np.float64)

ARGS = {
    'signal': signal_data,
    'target': 'pulse',
    'Fs': 30,
    'diff': True,
    'detrend_flag': True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (2,) — shape (2,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
interpolate_data Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
MayYoY/rPPGDIL/datasets/transforms.py
code served (permissive licence) · get_code("e2ff24c86102b414")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn.functional as F

torch.manual_seed(0)

# frames: expected to have shape (B, C, H, W) - will be unsqueezed to (1, B, C, H, W)
# For the function to work with trilinear interpolation, we need 5D tensor after unsqueeze
# frames.unsqueeze(0) makes it (1, B, C, H, W) which is correct for trilinear
frames = torch.randn(2, 3, 4, 5)  # (B=2, C=3, H=4, W=5)

# wave: expected to have shape (T,) - will be unsqueezed twice to (1, 1, T)
# for linear interpolation on the last dimension
wave = torch.randn(10)  # time dimension of length 10

# target_len: scalar integer for the target length
target_len = 15

ARGS = {
    "frames": frames,
    "wave": wave,
    "target_len": target_len,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
next_power_of_2 Ran
ran, honoured the contract we drafted
MayYoY/rPPGDIL/evaluate/postprocess.py
pointer only (licence: MIT); the stored copy get_code serves, from zzhang1987/truncated-matrix-power-iteration-for-differentiable-dag-learning, is pointer only (licence: NONE) · get_code("c9fc8e4585882848")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from zzhang1987/truncated-matrix-power-iteration-for-differentiable-dag-learning, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

normal_sampling Ran
ran, honoured the contract we drafted
MayYoY/rPPGDIL/loss_functions/frequent.py
code served (permissive licence) · get_code("54f858ca18dcc814")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import math

torch_imported = False
try:
    import torch
    torch_imported = True
except ImportError:
    pass

# Construct arguments for normal_sampling function
# The function computes probability density of a normal distribution
# Parameters:
# - mean: the mean of the distribution (numeric)
# - label_k: the point at which to evaluate (numeric)
# - std: standard deviation (numeric, default 1.0)

mean_value = 0.0
label_k_value = 1.5
std_value = 2.0

ARGS = {
    "mean": mean_value,
    "label_k": label_k_value,
    "std": std_value
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
read_video Ran
ran, honoured the contract we drafted
MayYoY/rPPGDIL/datasets/transforms.py
code served (permissive licence) · get_code("be9ee18a40895921")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 4, dtype family float, shape (3, 2, 4, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np
import os
import tempfile
from PIL import Image

torch.manual_seed(0)

# Create a temporary directory with dummy PNG files
temp_dir = tempfile.mkdtemp()

# Create 3 dummy PNG images (RGB format, 4x3 pixels for distinctness)
for i in range(0, 3):
    img_array = np.random.randint(0, 256, (4, 3, 3), dtype=np.uint8)
    img = Image.fromarray(img_array, 'RGB')
    img.save(os.path.join(temp_dir, f'{i}.png'))

ARGS = {
    'video_path': temp_dir,
    'start': 0,
    'end': 2
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, cv2, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (3, 2, 4, 3) — shape (3, 2, 4, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
resample_from_path Not yet run
unverified (harvested, no recorded run)
MayYoY/rPPGDIL/datasets/transforms.py
code served (permissive licence) · get_code("e5091edca4ac1339")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 0 later) and 7 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
next_power_of_2 2 other papers
2 dated earlier
2208.14571 (2022-08)
Truncated Matrix Power Iteration for Differentiable DAG Learning
2405.08363 (2024-05)
UnMarker: A Universal Attack on Defensive Image Watermarking

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Remote photoplethysmography (rPPG) has gained significant attention in recent years for its ability to extract physiological signals from facial videos. While existing rPPG measurement methods have shown satisfactory performance in intra-dataset and cross-dataset scenarios, they often overlook the incremental learning scenario, where training data is presented sequentially, resulting in the issue of catastrophic forgetting. Meanwhile, most existing class incremental learning approaches are unsuitable for rPPG measurement. In this paper, we present a novel method named ADDP to tackle continual learning for rPPG measurement. We first employ adapter to efficiently finetune the model on new tasks. Then we design domain prototypes that are more applicable to rPPG signal regression than commonly used class prototypes. Based on these prototypes, we propose a feature augmentation strategy to consolidate the past knowledge and an inference simplification strategy to convert potentially forgotten tasks into familiar ones for the model. To evaluate ADDP and enable fair comparisons, we create the first continual learning protocol for rPPG measurement. Comprehensive experiments demonstrate the effectiveness of our method for rPPG continual learning. Source code is available at https://github.com/MayYoY/rPPGDIL.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2407.13974")
get_code_for_paper("2407.13974")
have("2407.13974")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2407.13974.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2407.13974.svg)](https://syntology.ai/paper/2407.13974)

Connect an agent — have() is free.