CAMERA-P: Multi-Modal Imaging and Compression

• CAMERA-P denotes a structured micro-expert pruning approach in MoE models, achieving notable parameter reduction and speedups (up to 1.48×) at high pruning ratios.
• In coded optical sensing, CAMERA-P refers to a P-channel FDMA-CDMA configuration that enables simultaneous multi-channel imaging with enhanced speed and security.
• Additionally, CAMERA-P identifies specialized systems for single-pixel imaging, astronomical photometry, and Cherenkov detection, each with tailored design and calibration parameters.

Searching arXiv for recent and exact matches on “CAMERA-P” and the provided IDs to ground the article. arxiv_search(query="^^^^^^^^4CAMERA-P OR pCam6060 OR \4^^^^^^^^"Disturbance-free single-pixel imaging camera\"^^^^^^^^4 OR \4^^^^^^^^"Triple Coding Empowered FDMA-CDMA Mode High Security CAOS Camera\"^^^^^^^^4 OR \4^^^^^^^^"CAMERA: Multi-Matrix Joint Compression\"^^^^^^^^4 OR \4^^^^^^^^"Cherenkov Camera for the PBR mission\"", max_results=^^^^4 OR \4CAMERA-P OR pCam6060 OR \4^^^^, sort_by="submittedDate") > /tmp/arxiv_results.json && cat /tmp/arxiv_results.json

I found relevant arXiv records matching the provided usages of “CAMERA-P,” including the single-pixel imaging work, CAOS FDMA-CDMA mode, the pCam6^^^^4CAMERA-P OR pCam6060 OR \4^^^^6^^^^4CAMERA-P OR pCam6060 OR \4^^^^ photodetecting system, the MoE compression framework, and PBR Cherenkov-camera papers. CAMERA-P is a context-dependent designation used in several technically unrelated research literatures. In the available record, it can denote the structured pruning component of the CAMERA framework for mixture-of-experts compression, the pCam6^^^^4CAMERA-P OR pCam6060 OR \4^^^^6^^^^4CAMERA-P OR pCam6060 OR \4^^^^ photodetecting system optimized for astronomical photometry, a P-channel FDMA-CDMA operating mode of the CAOS camera, a disturbance-suppressed single-pixel imaging configuration associated with PRESERVED_PLACEHOLDER_^^^^4CAMERA-P OR pCam6060 OR \4^^^^ Bernoulli modulation, and, in some query conventions, fast SiPM-based Cherenkov camera packages for balloon-borne or ground-based high-energy astrophysics (&&&^^^^4CAMERA-P OR pCam6060 OR \4^^^^&&&, &&&^^^^4 OR \4^^^^&&&, &&&^^^^4 OR \4^^^^&&&, &&&^^^^4 OR \4^^^^&&&, Scotti et al., 19 Nov 2025, Bagheri et al., 2024). This suggests that CAMERA-P is best understood as a disambiguation term whose precise referent depends on subfield and paper-specific nomenclature.

^^^^4 OR \4^^^^. Scope and disambiguation

The term does not identify a single canonical instrument or algorithm. One paper explicitly defines CAMERA-P as a structured pruning method inside a broader MoE-compression framework. Other papers use closely related but not identical labels: one describes CAMERA-P as the pCam6^^^^4CAMERA-P OR pCam6060 OR \4^^^^6^^^^4CAMERA-P OR pCam6060 OR \4^^^^ astronomical photodetecting system, another explains that CAMERA-P is effectively the CAOS camera operated in a “P-channel” FDMA-CDMA configuration, and a PBR proceedings contribution notes that the query term appears to refer to the mission’s Cherenkov Camera (CC) rather than to an independent formal acronym. In the supplied single-pixel-imaging mapping, CAMERA-P is associated with the CSPI camera based on complementary detection and optimized Bernoulli modulation (&&&^^^^4CAMERA-P OR pCam6060 OR \4^^^^&&&, &&&^^^^4 OR \4^^^^&&&, &&&^^^^4 OR \4^^^^&&&, &&&^^^^4 OR \4^^^^&&&, Scotti et al., 19 Nov 2025, Bagheri et al., 2024).

Context	Referent	Core technical idea
MoE compression	CAMERA-P	Structured micro-expert pruning
Coded optical sensing	P-channel CAOS camera	FDMA-CDMA triple coding with PRESERVED_PLACEHOLDER_^^^^4 OR \4^^^^ frequency channels
Single-pixel imaging	CSPI / CAMERA-P mapping	Complementary detection with PRESERVED_PLACEHOLDER_^^^^4 OR \4^^^^ Bernoulli modulation
Astronomical photometry	pCam6^^^^4CAMERA-P OR pCam6060 OR \4^^^^6^^^^4CAMERA-P OR pCam6060 OR \4^^^^ / CAMERA-P	Large-format BSI CMOS HDR photodetecting system
High-energy astrophysics	PBR CC or modular SiPM camera	Fast Cherenkov detection with SiPM focal planes

This distribution matters because the meaning of the suffix “-P” changes across papers. In the MoE work it names a specific pruning operator; in the CAOS work it denotes the number of simultaneous FDMA channels; in pCam6^^^^4CAMERA-P OR pCam6060 OR \4^^^^6^^^^4CAMERA-P OR pCam6060 OR \4^^^^ it is part of the camera name; and in mission instrumentation papers it is an external query label mapped onto camera subsystems rather than the instrument’s own primary acronym.

^^^^4 OR \4^^^^. CAMERA-P as structured micro-expert pruning in MoE models

The most explicit formal definition appears in “CAMERA: Multi-Matrix Joint Compression for MoE Models via Micro-Expert Redundancy Analysis”, where CAMERA-P is the structured pruning instantiation of the CAMERA framework (&&&^^^^4CAMERA-P OR pCam6060 OR \4^^^^&&&). The paper’s central move is to replace the usual compression unit—expert, matrix, or channel—with the micro-expert, defined as the coupled computation associated with one hidden dimension across the expert FFN’s up, gate, and down projections. In the paper’s notation,

PRESERVED_PLACEHOLDER_^^^^4 OR \4^^^^

This induces a decomposition of the MoE layer as a mixture of micro-experts,

$$\mathbf{y} = \sum_{i=1}^{N_e} \phi_i \mathbf{w}_i^{\mathrm{down}}, \quad \phi_i = A_i(\mathbf{x})\cdot \sigma(\mathbf{w}_i^{\mathrm{gate}}\mathbf{x})\cdot \mathbf{w}_i^{\mathrm{up}}\mathbf{x}.$$

The redundancy analysis is posed over a calibration set $\{(\mathbf{x}_i,\mathbf{y}_i)\}_{i=1}^n$ through

$\mathbf{Y} = \mathbf{\Phi}\mathbf{W},$

followed by a column-subset selection objective,

$$\min_{S\subset [N_e],\,|S|=m}\left\|\mathbf{Y}-\mathbf{\Phi}_{:,S}\mathbf{W}_{S,:}\right\|_F^2.$$

The exact selection is framed as NP-hard, so the method ranks micro-experts by a decoding-time energy score. The paper first motivates

$$\epsilon_{\mathrm{sup}} = \sum_{i\in S^C}\|\mathbf{\Phi}_{:,i}\|_2^2 \|\mathbf{w}_i\|_2^2,$$

then defines

$$\mathcal{E}_i = \left[(1-\alpha)\|\mathbf{\Phi}_{:,i}\|_2^2 + \alpha\|\mathbf{\Phi}_{:,i}\|_\infty^2\right]\cdot \|\mathbf{w}_i\|_2^2.$$

Lower energy indicates greater redundancy.

CAMERA-P uses this ranking for joint cross-matrix pruning. For pruning ratio PRESERVED_PLACEHOLDER_^^^^4 OR \4CAMERA-P OR pCam6060 OR \4^^^^, it removes the lowest-energy PRESERVED_PLACEHOLDER_^^^^4 OR \4 OR \4^^^^ fraction of micro-experts in each MoE layer and deletes the corresponding row in PRESERVED_PLACEHOLDER_^^^^4 OR \4 OR \4^^^^, row in PRESERVED_PLACEHOLDER_^^^^4 OR \4 OR \4^^^^, and matching column in PRESERVED_PLACEHOLDER_^^^^4 OR \44^^^^. The layerwise update is written as

PRESERVED_PLACEHOLDER_^^^^4 OR \4^^^^5

Because the pruning unit is structurally coupled across all three matrices, the resulting FFN is reduced in width without introducing unstructured sparsity.

The empirical evaluation centers on Deepseek-MoE-^^^^4 OR \4^^^^6B, Qwen^^^^4 OR \4^^^^-57B-A^^^^4 OR \44^^^^B, and Qwen^^^^4 OR \4^^^^-^^^^4 OR \4CAMERA-P OR pCam6060 OR \4^^^^B-A^^^^4 OR \4^^^^B, with calibration on Wikitext^^^^4 OR \4^^^^, ^^^^4 OR \4 OR \4^^^^8 sequences, and ^^^^4 OR \4CAMERA-P OR pCam6060 OR \44^^^^8 tokens. Reported pruning ratios are ^^^^4 OR \4CAMERA-P OR pCam6060 OR \4^^^^%, 4^^^^4CAMERA-P OR pCam6060 OR \4^^^^%, and 6^^^^4CAMERA-P OR pCam6060 OR \4^^^^%. The method consistently exceeds NAEE and PRESERVED_PLACEHOLDER_^^^^4 OR \4^^^^6-MoE on average zero-shot accuracy, with particularly strong behavior at aggressive pruning. On Qwen^^^^4 OR \4^^^^-57B-A^^^^4 OR \44^^^^B at 6^^^^4CAMERA-P OR pCam6060 OR \4^^^^% pruning, the reported average scores are 5^^^^4 OR \4^^^^.4^^^^4CAMERA-P OR pCam6060 OR \4^^^^ for NAEE, 56.^^^^4 OR \4 OR \4^^^^ for PRESERVED_PLACEHOLDER_^^^^4 OR \4^^^^7-MoE, and 65.^^^^4 OR \4^^^^7 for CAMERA-P. The paper also reports that complete micro-expert analysis of Qwen^^^^4 OR \4^^^^-57B-A^^^^4 OR \44^^^^B takes less than 5 minutes on a single NVIDIA A^^^^4 OR \4CAMERA-P OR pCam6060 OR \4CAMERA-P OR pCam6060 OR \4^^^^-4^^^^4CAMERA-P OR pCam6060 OR \4^^^^GB GPU, and that pruning completes in about ^^^^4CAMERA-P OR pCam6060 OR \4^^^^.^^^^4 OR \4^^^^ GPU hours. This suggests that CAMERA-P is designed not only for parameter reduction but for actual structural speedup, which the paper quantifies on Deepseek-MoE-^^^^4 OR \4^^^^6B as PRESERVED_PLACEHOLDER_^^^^4 OR \4^^^^8, PRESERVED_PLACEHOLDER_^^^^4 OR \4^^^^9, and PRESERVED_PLACEHOLDER_^^^^4 OR \4CAMERA-P OR pCam6060 OR \4^^^^ at ^^^^4 OR \4CAMERA-P OR pCam6060 OR \4^^^^%, 4^^^^4CAMERA-P OR pCam6060 OR \4^^^^%, and 6^^^^4CAMERA-P OR pCam6060 OR \4^^^^% pruning.

^^^^4 OR \4^^^^. CAMERA-P as a P-channel FDMA-CDMA CAOS camera mode

In the CAOS literature, CAMERA-P is not introduced as a separate formal acronym; rather, it is effectively the CAOS camera operated in a “P-channel” FDMA-CDMA configuration (&&&^^^^4 OR \4^^^^&&&). The underlying platform is the CAOS (Coded Access Optical Sensor) camera, a programmable “thinking camera” based on a DMD spatial modulator and point detectors. The hybrid mode combines three coding dimensions: space coding, time coding with Walsh CDMA, and frequency coding with FDMA carriers. The paper describes this as a space-time-frequency triple coding design.

The encoded photocurrent is modeled as

PRESERVED_PLACEHOLDER_^^^^4 OR \4 OR \4^^^^

with

PRESERVED_PLACEHOLDER_^^^^4 OR \4 OR \4^^^^

Here PRESERVED_PLACEHOLDER_^^^^4 OR \4 OR \4^^^^ is the irradiance of the PRESERVED_PLACEHOLDER_^^^^4 OR \44^^^^-th CAOS pixel, PRESERVED_PLACEHOLDER_^^^^4 OR \4^^^^5 is the Walsh CDMA code, and PRESERVED_PLACEHOLDER_^^^^4 OR \4^^^^6 is the FDMA frequency assignment. The decoding chain performs FFT analysis during each CDMA bit duration PRESERVED_PLACEHOLDER_^^^^4 OR \4^^^^7, assembles the PRESERVED_PLACEHOLDER_^^^^4 OR \4^^^^8 spectral peaks over PRESERVED_PLACEHOLDER_^^^^4 OR \4^^^^9 bits, correlates them with the PRESERVED_PLACEHOLDER_^^^^4 OR \4CAMERA-P OR pCam6060 OR \4^^^^ Walsh code sequences, and maps the recovered irradiances back to the spatial grid. The key counting relation is

PRESERVED_PLACEHOLDER_^^^^4 OR \4 OR \4^^^^

where PRESERVED_PLACEHOLDER_^^^^4 OR \4 OR \4^^^^ is the total number of CAOS pixels.

The parameter PRESERVED_PLACEHOLDER_^^^^4 OR \4 OR \4^^^^ denotes the number of simultaneous FDMA channels, and the paper states that the demonstrated FDMA-CDMA mode operates PRESERVED_PLACEHOLDER_^^^^4 OR \44^^^^ times faster than the equivalent linear HDR FM-CDMA mode. The example given is PRESERVED_PLACEHOLDER_^^^^4 OR \4^^^^5, PRESERVED_PLACEHOLDER_^^^^4 OR \4^^^^6, PRESERVED_PLACEHOLDER_^^^^4 OR \4^^^^7, and Walsh length PRESERVED_PLACEHOLDER_^^^^4 OR \4^^^^8 instead of ^^^^4 OR \4CAMERA-P OR pCam6060 OR \44^^^^8, implying an 8-fold reduction in encoding time. The stated benefits are high security, linear HDR, high SNR, and higher speed. The HDR claim is tied to FFT/DSP-based spectrum analysis, while the SNR claim is tied to simultaneous multi-pixel detection in the CDMA mode.

The active-mode demonstration uses PRESERVED_PLACEHOLDER_^^^^4 OR \4^^^^9 LEDs driven at $$\mathbf{y} = \sum_{i=1}^{N_e} \phi_i \mathbf{w}_i^{\mathrm{down}}, \quad \phi_i = A_i(\mathbf{x})\cdot \sigma(\mathbf{w}_i^{\mathrm{gate}}\mathbf{x})\cdot \mathbf{w}_i^{\mathrm{up}}\mathbf{x}.$$^^^^4CAMERA-P OR pCam6060 OR \4^^^^ kHz, $$\mathbf{y} = \sum_{i=1}^{N_e} \phi_i \mathbf{w}_i^{\mathrm{down}}, \quad \phi_i = A_i(\mathbf{x})\cdot \sigma(\mathbf{w}_i^{\mathrm{gate}}\mathbf{x})\cdot \mathbf{w}_i^{\mathrm{up}}\mathbf{x}.$$^^^^4 OR \4^^^^ kHz, and $$\mathbf{y} = \sum_{i=1}^{N_e} \phi_i \mathbf{w}_i^{\mathrm{down}}, \quad \phi_i = A_i(\mathbf{x})\cdot \sigma(\mathbf{w}_i^{\mathrm{gate}}\mathbf{x})\cdot \mathbf{w}_i^{\mathrm{up}}\mathbf{x}.$$^^^^4 OR \4^^^^ kHz, each with distinct optical spectral content. The paper emphasizes that this permits simultaneous capture of $$\mathbf{y} = \sum_{i=1}^{N_e} \phi_i \mathbf{w}_i^{\mathrm{down}}, \quad \phi_i = A_i(\mathbf{x})\cdot \sigma(\mathbf{w}_i^{\mathrm{gate}}\mathbf{x})\cdot \mathbf{w}_i^{\mathrm{up}}\mathbf{x}.$$^^^^4 OR \4^^^^ images without time-multiplexed slots and without a tunable optical filter. In that sense, CAMERA-P denotes a parallel active sensing camera whose rate and coding burden scale with the number of FDMA channels, rather than a separate camera body.

4. CAMERA-P as disturbance-free single-pixel imaging with $$\mathbf{y} = \sum_{i=1}^{N_e} \phi_i \mathbf{w}_i^{\mathrm{down}}, \quad \phi_i = A_i(\mathbf{x})\cdot \sigma(\mathbf{w}_i^{\mathrm{gate}}\mathbf{x})\cdot \mathbf{w}_i^{\mathrm{up}}\mathbf{x}.$$4

In the supplied single-pixel-imaging mapping, CAMERA-P corresponds to the single-pixel imaging camera based on complementary detection and optimized Bernoulli modulation, called CSPI in the paper itself (&&&^^^^4 OR \4^^^^&&&). The architecture uses a DMD to project binary coded patterns, records two complementary measurements simultaneously, and subtracts the detector outputs. One detector sees $$\mathbf{y} = \sum_{i=1}^{N_e} \phi_i \mathbf{w}_i^{\mathrm{down}}, \quad \phi_i = A_i(\mathbf{x})\cdot \sigma(\mathbf{w}_i^{\mathrm{gate}}\mathbf{x})\cdot \mathbf{w}_i^{\mathrm{up}}\mathbf{x}.$$5, the other sees $$\mathbf{y} = \sum_{i=1}^{N_e} \phi_i \mathbf{w}_i^{\mathrm{down}}, \quad \phi_i = A_i(\mathbf{x})\cdot \sigma(\mathbf{w}_i^{\mathrm{gate}}\mathbf{x})\cdot \mathbf{w}_i^{\mathrm{up}}\mathbf{x}.$$6. The recorded intensities are

$$\mathbf{y} = \sum_{i=1}^{N_e} \phi_i \mathbf{w}_i^{\mathrm{down}}, \quad \phi_i = A_i(\mathbf{x})\cdot \sigma(\mathbf{w}_i^{\mathrm{gate}}\mathbf{x})\cdot \mathbf{w}_i^{\mathrm{up}}\mathbf{x}.$$7

$$\mathbf{y} = \sum_{i=1}^{N_e} \phi_i \mathbf{w}_i^{\mathrm{down}}, \quad \phi_i = A_i(\mathbf{x})\cdot \sigma(\mathbf{w}_i^{\mathrm{gate}}\mathbf{x})\cdot \mathbf{w}_i^{\mathrm{up}}\mathbf{x}.$$8

and the differential CSPI measurement is

$$\mathbf{y} = \sum_{i=1}^{N_e} \phi_i \mathbf{w}_i^{\mathrm{down}}, \quad \phi_i = A_i(\mathbf{x})\cdot \sigma(\mathbf{w}_i^{\mathrm{gate}}\mathbf{x})\cdot \mathbf{w}_i^{\mathrm{up}}\mathbf{x}.$$9

This yields

$\{(\mathbf{x}_i,\mathbf{y}_i)\}_{i=1}^n$^^^^4CAMERA-P OR pCam6060 OR \4^^^^

The disturbance-cancellation mechanism depends on Bernoulli-distributed binary patterns with

$\{(\mathbf{x}_i,\mathbf{y}_i)\}_{i=1}^n$^^^^4 OR \4^^^^

When $\{(\mathbf{x}_i,\mathbf{y}_i)\}_{i=1}^n$^^^^4 OR \4^^^^, the mean is exactly $\{(\mathbf{x}_i,\mathbf{y}_i)\}_{i=1}^n$^^^^4 OR \4^^^^, and in the ideal balanced case

$\{(\mathbf{x}_i,\mathbf{y}_i)\}_{i=1}^n$4

Under the paper’s stated assumption that the disturbance light is spatially uniform on the DMD plane but may vary arbitrarily from one measurement to the next, the term involving $\{(\mathbf{x}_i,\mathbf{y}_i)\}_{i=1}^n$5 vanishes. The paper therefore describes the camera as disturbance-free under those conditions. The reconstruction retains the standard correlation SPI framework:

$\{(\mathbf{x}_i,\mathbf{y}_i)\}_{i=1}^n$6

$\{(\mathbf{x}_i,\mathbf{y}_i)\}_{i=1}^n$7

The paper compares standard SPI, differential SPI (DSPI), and CSPI. DSPI is given as

$\{(\mathbf{x}_i,\mathbf{y}_i)\}_{i=1}^n$8

with $\{(\mathbf{x}_i,\mathbf{y}_i)\}_{i=1}^n$9. The reported simulations show that DSPI can mitigate slowly varying background, but fails when the disturbance fluctuates rapidly or strongly relative to the signal, whereas CSPI remains effective even when disturbance intensity is rapidly varying, random, and much stronger than the target return. With sinusoidal disturbance and irradiation SNR $\mathbf{Y} = \mathbf{\Phi}\mathbf{W},$^^^^4CAMERA-P OR pCam6060 OR \4^^^^ dB, the reported correlation coefficient $\mathbf{Y} = \mathbf{\Phi}\mathbf{W},$^^^^4 OR \4^^^^ is ^^^^4CAMERA-P OR pCam6060 OR \4^^^^.659 for DSPI and $\mathbf{Y} = \mathbf{\Phi}\mathbf{W},$^^^^4 OR \4^^^^ for CSPI. The paper also reports explicit tests for $\mathbf{Y} = \mathbf{\Phi}\mathbf{W},$^^^^4 OR \4^^^^, $\mathbf{Y} = \mathbf{\Phi}\mathbf{W},$4, and $\mathbf{Y} = \mathbf{\Phi}\mathbf{W},$5, with best reconstruction at $\mathbf{Y} = \mathbf{\Phi}\mathbf{W},$6. For a $\mathbf{Y} = \mathbf{\Phi}\mathbf{W},$7 case using a Hadamard-coded arrangement of Bernoulli patterns, removal of the first all-ones pattern leaves $\mathbf{Y} = \mathbf{\Phi}\mathbf{W},$8 measurements. A plausible implication is that, within the paper’s assumptions, CAMERA-P denotes a pattern-statistics-based cancellation scheme rather than ordinary background subtraction.

5. CAMERA-P as the pCam6^^^^4CAMERA-P OR pCam6060 OR \4^^^^6^^^^4CAMERA-P OR pCam6060 OR \4^^^^ astronomical photodetecting system

In astronomical detector development, CAMERA-P directly refers to the pCam6^^^^4CAMERA-P OR pCam6060 OR \4^^^^6^^^^4CAMERA-P OR pCam6060 OR \4^^^^ photodetecting system developed at the Special Astrophysical Observatory of the Russian Academy of Sciences and optimized for photometric observations (&&&^^^^4 OR \4^^^^&&&). The system is built around the GSENSE6^^^^4CAMERA-P OR pCam6060 OR \4^^^^6^^^^4CAMERA-P OR pCam6060 OR \4^^^^BSI back-illuminated CMOS detector with $\mathbf{Y} = \mathbf{\Phi}\mathbf{W},$9 active pixels and $$\min_{S\subset [N_e],\,|S|=m}\left\|\mathbf{Y}-\mathbf{\Phi}_{:,S}\mathbf{W}_{S,:}\right\|_F^2.$$^^^^4CAMERA-P OR pCam6060 OR \4^^^^ pixel size. The paper reports a full-frame readout rate of ^^^^4 OR \4 OR \4^^^^ fps, communication via a fiber-optic line at distances of up to 5^^^^4CAMERA-P OR pCam6060 OR \4^^^^ m, and real-time recording of video data to the computer hard drive.

The detector’s reported spectral response is ^^^^4 OR \4CAMERA-P OR pCam6060 OR \4CAMERA-P OR pCam6060 OR \4^^^^–^^^^4 OR \4CAMERA-P OR pCam6060 OR \44CAMERA-P OR pCam6060 OR \4^^^^ nm, with minimum QE ^^^^4 OR \4CAMERA-P OR pCam6060 OR \4^^^^%, maximum QE 95% at 58^^^^4CAMERA-P OR pCam6060 OR \4^^^^ nm, and QE 58% at 85^^^^4CAMERA-P OR pCam6060 OR \4^^^^ nm. The BSI architecture is emphasized because, unlike front-illuminated devices, it does not show the long-term residual bulk image effect characteristic of many FSI detectors. The paper does note a small residual lag after saturation: after reset, about ^^^^4 OR \4^^^^ e$$\min_{S\subset [N_e],\,|S|=m}\left\|\mathbf{Y}-\mathbf{\Phi}_{:,S}\mathbf{W}_{S,:}\right\|_F^2.$$^^^^4 OR \4^^^^ remain in saturated pixels in the next frame. This is one of the reasons the system is presented as particularly suitable for faint-object photometry and long-exposure observation methods.

A distinctive feature is the simultaneous readout through two ^^^^4 OR \4 OR \4^^^^-bit video channels with different gain settings and their controller-level combination into a single ^^^^4 OR \4^^^^6-bit frame. The readout modes are LG, HG, and HDR. The paper describes the HDR method as using multiplicative and additive coefficients derived from measured transfer characteristics so that the combined response is linear and shows no gain shift or dispersion shift at the junction between channels. The reported gains are $$\min_{S\subset [N_e],\,|S|=m}\left\|\mathbf{Y}-\mathbf{\Phi}_{:,S}\mathbf{W}_{S,:}\right\|_F^2.$$^^^^4 OR \4^^^^ for LG, $$\min_{S\subset [N_e],\,|S|=m}\left\|\mathbf{Y}-\mathbf{\Phi}_{:,S}\mathbf{W}_{S,:}\right\|_F^2.$$^^^^4 OR \4^^^^ for HG, and $$\min_{S\subset [N_e],\,|S|=m}\left\|\mathbf{Y}-\mathbf{\Phi}_{:,S}\mathbf{W}_{S,:}\right\|_F^2.$$4 for HDR. The corresponding readout noises are ^^^^4 OR \4 OR \4^^^^.6 e$$\min_{S\subset [N_e],\,|S|=m}\left\|\mathbf{Y}-\mathbf{\Phi}_{:,S}\mathbf{W}_{S,:}\right\|_F^2.$$5 rms, ^^^^4 OR \4^^^^.47 e$$\min_{S\subset [N_e],\,|S|=m}\left\|\mathbf{Y}-\mathbf{\Phi}_{:,S}\mathbf{W}_{S,:}\right\|_F^2.$$6 rms, and ^^^^4 OR \4^^^^.^^^^4 OR \4^^^^9 e$$\min_{S\subset [N_e],\,|S|=m}\left\|\mathbf{Y}-\mathbf{\Phi}_{:,S}\mathbf{W}_{S,:}\right\|_F^2.$$7 rms; the full well capacities are 9^^^^4 OR \4^^^^,^^^^4 OR \4CAMERA-P OR pCam6060 OR \4CAMERA-P OR pCam6060 OR \4^^^^ e$$\min_{S\subset [N_e],\,|S|=m}\left\|\mathbf{Y}-\mathbf{\Phi}_{:,S}\mathbf{W}_{S,:}\right\|_F^2.$$8, 9,^^^^4 OR \4CAMERA-P OR pCam6060 OR \4CAMERA-P OR pCam6060 OR \4^^^^ e$$\min_{S\subset [N_e],\,|S|=m}\left\|\mathbf{Y}-\mathbf{\Phi}_{:,S}\mathbf{W}_{S,:}\right\|_F^2.$$9, and 9^^^^4 OR \4^^^^,5^^^^4CAMERA-P OR pCam6060 OR \4CAMERA-P OR pCam6060 OR \4^^^^ e$$\epsilon_{\mathrm{sup}} = \sum_{i\in S^C}\|\mathbf{\Phi}_{:,i}\|_2^2 \|\mathbf{w}_i\|_2^2,$$^^^^4CAMERA-P OR pCam6060 OR \4^^^^; and the dynamic ranges are 7^^^^4 OR \4^^^^.9 dB, 68.6 dB, and 89.^^^^4 OR \4^^^^ dB.

The photometric characterization further reports non-linearity of ^^^^4CAMERA-P OR pCam6060 OR \4^^^^.6^^^^4 OR \4^^^^% in LG, ^^^^4CAMERA-P OR pCam6060 OR \4^^^^.8^^^^4CAMERA-P OR pCam6060 OR \4^^^^% in HG, and ^^^^4CAMERA-P OR pCam6060 OR \4^^^^.69% in HDR; photoresponse non-uniformity ^^^^4CAMERA-P OR pCam6060 OR \4^^^^.5%; gain instability ^^^^4CAMERA-P OR pCam6060 OR \4^^^^.^^^^4CAMERA-P OR pCam6060 OR \4^^^^64%; image lag ^^^^4 OR \4^^^^–4 e$$\epsilon_{\mathrm{sup}} = \sum_{i\in S^C}\|\mathbf{\Phi}_{:,i}\|_2^2 \|\mathbf{w}_i\|_2^2,$$^^^^4 OR \4^^^^/pixel; and dark current ^^^^4CAMERA-P OR pCam6060 OR \4^^^^.^^^^4 OR \4^^^^–^^^^4CAMERA-P OR pCam6060 OR \4^^^^.^^^^4 OR \4^^^^ e$$\epsilon_{\mathrm{sup}} = \sum_{i\in S^C}\|\mathbf{\Phi}_{:,i}\|_2^2 \|\mathbf{w}_i\|_2^2,$$^^^^4 OR \4^^^^/s/pixel. Thermal stability is critical because the gain dependence is reported as

$$\epsilon_{\mathrm{sup}} = \sum_{i\in S^C}\|\mathbf{\Phi}_{:,i}\|_2^2 \|\mathbf{w}_i\|_2^2,$$^^^^4 OR \4^^^^

The controller therefore stabilizes the detector temperature to $$\epsilon_{\mathrm{sup}} = \sum_{i\in S^C}\|\mathbf{\Phi}_{:,i}\|_2^2 \|\mathbf{w}_i\|_2^2,$$4C, which keeps gain instability below ^^^^4CAMERA-P OR pCam6060 OR \4^^^^.^^^^4CAMERA-P OR pCam6060 OR \4^^^^64%. Cooling uses two-stage Peltier elements, and the operating temperature can be brought to $$\epsilon_{\mathrm{sup}} = \sum_{i\in S^C}\|\mathbf{\Phi}_{:,i}\|_2^2 \|\mathbf{w}_i\|_2^2,$$5C below ambient radiator temperature. The body dimensions are ^^^^4 OR \4^^^^9^^^^4CAMERA-P OR pCam6060 OR \4^^^^ × ^^^^4 OR \4^^^^9^^^^4CAMERA-P OR pCam6060 OR \4^^^^ × ^^^^4 OR \4^^^^7^^^^4CAMERA-P OR pCam6060 OR \4^^^^ mm, and the system is described as moisture-proof. Within the astronomical context, CAMERA-P therefore designates a large-format, dual-gain, BSI-CMOS photometric camera emphasizing QE, readout-noise suppression, and HDR linearity.

6. CAMERA-P in high-energy astrophysics instrumentation

A separate usage associates CAMERA-P with fast Cherenkov-camera instrumentation for high-energy air showers. In the PBR proceedings contribution, the instrument is consistently described as the Cherenkov Camera (CC), and the supplied mapping states that the query term appears to refer to this same package on board POEMMA-Balloon with Radio (PBR) (Scotti et al., 19 Nov 2025). The CC is a ^^^^4 OR \4CAMERA-P OR pCam6060 OR \44^^^^8-pixel SiPM camera operating over ^^^^4 OR \4 OR \4CAMERA-P OR pCam6060 OR \4^^^^–9^^^^4CAMERA-P OR pCam6060 OR \4CAMERA-P OR pCam6060 OR \4^^^^ nm with ^^^^4 OR \4CAMERA-P OR pCam6060 OR \4^^^^ ns integration time. Its focal surface consists of four rows of 8 SiPM arrays, each array containing $$\epsilon_{\mathrm{sup}} = \sum_{i\in S^C}\|\mathbf{\Phi}_{:,i}\|_2^2 \|\mathbf{w}_i\|_2^2,$$6 channels, yielding

$$\epsilon_{\mathrm{sup}} = \sum_{i\in S^C}\|\mathbf{\Phi}_{:,i}\|_2^2 \|\mathbf{w}_i\|_2^2,$$7

pixels. The reported field of view is $$\epsilon_{\mathrm{sup}} = \sum_{i\in S^C}\|\mathbf{\Phi}_{:,i}\|_2^2 \|\mathbf{w}_i\|_2^2,$$8 with pixel angular scale of about $$\epsilon_{\mathrm{sup}} = \sum_{i\in S^C}\|\mathbf{\Phi}_{:,i}\|_2^2 \|\mathbf{w}_i\|_2^2,$$9 per pixel. The optical system is a Schmidt-type telescope with ^^^^4 OR \4^^^^.^^^^4 OR \4^^^^ m aperture, approximately ^^^^4 OR \4^^^^.6 m radius of curvature, an aspheric corrector plate made of UV-transparent PMMA, and a segmented primary mirror of ^^^^4 OR \4 OR \4^^^^ vacuum-slumped borosilicate glass elements. A distinctive element is the bi-focal optical design, implemented with a PMMA ^^^^4 OR \4^^^^D prism array mounted ^^^^4 OR \4CAMERA-P OR pCam6060 OR \4CAMERA-P OR pCam6060 OR \4^^^^ mm in front of the focal plane, which splits incoming light into two spatially separated spots and supports a coincidence-based trigger. The readout is based on the MIZAR ASIC, which handles 64 channels, samples at ^^^^4 OR \4CAMERA-P OR pCam6060 OR \4CAMERA-P OR pCam6060 OR \4^^^^ MHz, provides ^^^^4 OR \4^^^^56 memory cells, and offers a single-slope ADC with programmable resolution of 7–^^^^4 OR \4 OR \4^^^^ bits. The trigger requires that at least two pixels exceed the thresholds.

A different but related paper uses CAMERA-P for the modular SiPM camera and readout system developed for the Trinity Demonstrator and the EUSO-SPB^^^^4 OR \4^^^^ Cherenkov Telescope (Bagheri et al., 2024). Here the camera is optimized for Earth-skimming PeV–EeV tau neutrinos observed with the imaging atmospheric Cherenkov technique. Two versions are reported: a ^^^^4 OR \4^^^^56-pixel camera for Trinity using Hamamatsu S^^^^4 OR \44 OR \4^^^^6^^^^4 OR \4^^^^-6^^^^4CAMERA-P OR pCam6060 OR \4submittedDate4CAMERA-P OR pCam6060 OR \4^^^^HS SiPMs, and a 5^^^^4 OR \4 OR \4^^^^-pixel camera for EUSO-SPB^^^^4 OR \4^^^^ using Hamamatsu S^^^^4 OR \44submittedDate4 OR \4 OR \4^^^^-6^^^^4CAMERA-P OR pCam6060 OR \4submittedDate4CAMERA-P OR pCam6060 OR \4^^^^AN SiPMs. The front end is built around the eMUSIC ASIC, and the signals are sampled and digitized with the AGET system at ^^^^4 OR \4CAMERA-P OR pCam6060 OR \4CAMERA-P OR pCam6060 OR \4^^^^ MS/s and ^^^^4 OR \4 OR \4^^^^-bit resolution. Both cameras are liquid-cooled. Each camera module contains a 4×4 matrix of SiPMs, front-end Sensor Interface and Amplification Board (SIAB) electronics, and, where required, adaptor boards for curved focal planes. The AGET chain uses a 5^^^^4 OR \4 OR \4^^^^-cell buffer depth and yields a 5.^^^^4 OR \4 OR \4^^^^ μs trace length.

The quantitative characterization is unusually detailed. The selected operating point is where PDE reaches 9^^^^4CAMERA-P OR pCam6060 OR \4^^^^% of its maximum, corresponding to about 9% relative overvoltage for the reference device at room temperature, or about 44 V bias. The paper reports breakdown-voltage temperature dependence of ^^^^4 OR \4^^^^7 mV/°C, direct optical crosstalk of around ~^^^^4 OR \4^^^^%, afterpulsing of about 5%, and a recovery time constant of about ^^^^4 OR \4CAMERA-P OR pCam6060 OR \4CAMERA-P OR pCam6060 OR \4^^^^ ns. Signal-chain calibration gives $$\mathcal{E}_i = \left[(1-\alpha)\|\mathbf{\Phi}_{:,i}\|_2^2 + \alpha\|\mathbf{\Phi}_{:,i}\|_\infty^2\right]\cdot \|\mathbf{w}_i\|_2^2.$$^^^^4CAMERA-P OR pCam6060 OR \4^^^^ digital counts per photoelectron, with linearity up to about ^^^^4 OR \4CAMERA-P OR pCam6060 OR \4CAMERA-P OR pCam6060 OR \4^^^^ photoelectrons. The raw pulses are slowed for AGET compatibility by a third-order low-pass Butterworth filter with ^^^^4 OR \4^^^^5 MHz cutoff, producing rise time of about ^^^^4 OR \4CAMERA-P OR pCam6060 OR \4^^^^ ns and output pulses with ~^^^^4 OR \4CAMERA-P OR pCam6060 OR \4^^^^ ns FWHM. Flatfielding is performed through per-pixel bias trimming according to

$$\mathcal{E}_i = \left[(1-\alpha)\|\mathbf{\Phi}_{:,i}\|_2^2 + \alpha\|\mathbf{\Phi}_{:,i}\|_\infty^2\right]\cdot \|\mathbf{w}_i\|_2^2.$$^^^^4 OR \4^^^^

and after flatfielding the reported pixel-response standard deviation is about ^^^^4CAMERA-P OR pCam6060 OR \4^^^^.^^^^4CAMERA-P OR pCam6060 OR \4^^^^5.

Taken together, these high-energy uses show that CAMERA-P can denote either a mission-specific Cherenkov focal-plane package or a modular SiPM readout architecture for fast air-shower imaging. The shared technical pattern is a focus on fast timing, SiPM pixelation, and coincidence-capable electronics, but the papers describe distinct camera systems rather than a single unified platform.